Академический Документы
Профессиональный Документы
Культура Документы
Human Development
1 2 3 4 :: 14 13 12 11
This volume is a product of the staff of the International Bank for Reconstruction and
Development / The World Bank. The findings, interpretations, and conclusions expressed
in this volume do not necessarily reflect the views of the Executive Directors of The World
Bank or the governments they represent.
The World Bank does not guarantee the accuracy of the data included in this work. The
boundaries, colors, denominations, and other information shown on any map in this work
do not imply any judgement on the part of The World Bank concerning the legal status of
any territory or the endorsement or acceptance of such boundaries.
For permission to photocopy or reprint any part of this work, please send a request with
complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive,
Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www
.copyright.com.
All other queries on rights and licenses, including subsidiary rights, should be addressed to
the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433,
USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org.
ISBN: 978-0-8213-8096-3
eISBN: 978-0-8213-8097-0
DOI: 10.1596/978-0-8213-8096-3
Cover photo: The graduating class of 2007, Alexandru Ioan Cuza University, Iaşi, Romania,
march to their commencement ceremony. © Mediafax Foto/Liviu Chirica.
Contents
Foreword xv
Acknowledgments xix
Abbreviations xxi
Overview 1
The Skills Challenge 2
Why Are Skills an Emerging Problem if
Education Systems Are Delivering? 4
Priority Areas for Action 7
Managing Education Systems for Results 9
Build the Foundations of Adult Learning Systems 14
Conclusion 14
Notes 15
References 225
viii Contents
Boxes
2.1 Defining “Skills,” “Competencies,” and Other Terms 49
2.2 Testing for Reading Competency in the Programme
for International Student Assessment (PISA) 55
2.3 Turkey: Many Students Fail to Acquire Even the
Most Basic Proficiencies 57
2.4 Russia: Building Higher-Order Skills Is Proving Difficult 59
2.5 The History of Vocational Education in the ECA Region 64
2.6 Two Forms of Adult Learning 69
3.1 The Value of Standardized External Student
Assessments in ECA Countries 94
3.2 Detailed Regulations Set Norms Even for
Education Facilities in the ECA Region 99
4.1 Using Data to Measure Gaps and Design Better
Policies: Three Examples 118
4.2 Eliminating Class Size Norms in Denmark 124
4.3 Rethinking Secondary Education 125
4.4 Per Student Funding Formula: Recognizing the
Varied Costs of Providing Education 128
4.5 Rationalizing Class Sizes in Bulgaria 133
4.6 Developing New Teacher Policies 135
5.1 The Difficulty of Measuring Competencies at
the Tertiary Level 142
5.2 Decentralization of the University Sector in Romania 146
5.3 Introducing Businesslike Leadership and
Management in Lithuanian Universities 149
5.4 Using External Watchdogs to Shed Light
on Integrity Problems 155
5.5 Two Approaches to Strengthening Accountability 157
6.1 Lifelong Learning Strategy of the Czech Republic 167
6.2 Policy Tools for Advancing Adult Learning in the
United Kingdom 169
6.3 Market Failures that Impede Adult Learning 176
6.4 Supporting Small and Medium Enterprises to
Participate in Training 181
6.5 Second-Chance and Remedial Education 184
7.1 Higher-Order Skills for the World of Work in
the 21st Century 199
7.2 A Large Proportion of Students Are Failing 201
Contents ix
Figures
O.1 Analysis of Reading Competency of 15-Year-Old
Students on the PISA 2009 3
O.2 Distribution of Firms in ECA Region that Consider
Worker Skills a “Major” or “Very Severe”
Constraint, 2008 4
O.3 Primary School Student-Teacher Ratios in ECA
Compared to Other Regions of the World, 1990–2008 8
O.4 Status of Measuring and Using Data on Student
Learning Outcomes in the ECA Region, 2009 10
1.1 Trends in Routine and Nonroutine Tasks of U.S.
Labor Force, 1960–2002 19
1.2 Worker Responses to Survey on Working Conditions
in Europe, 2005 20
1.3 Job Creation and Destruction Rates for Selected
Occupations in Georgia, 2007 22
1.4 Job Creation and Destruction Rates for Selected
Occupations in Ukraine, 2007 23
1.5 Changes in Occupational Share of Total Employment
in Selected ECA Countries, Various Years 24
1.6 Yearly Outflow Rates from Employment to Jobs
by Occupation in Two Subregions of ECA, 2006 27
1.7 Unemployment-to-Vacancy Ratio by Occupation in
Croatia, the Czech Republic, and Ukraine, Various Years 28
1.8 Labor Shortages and Surpluses by Level of
Educational Attainment, Selected ECA Countries 29
1.9 Returns to Schooling in Selected ECA Transition
Countries, Various Years 32
1.10 Wage Premia by Level of Education in
Hungary, 1986–2004 33
1.11 Wage Premia by Education Level in the Russian
Federation and Turkey 34
1.12 Premia by Occupation Relative to Elementary
Occupations in Various ECA Countries 35
1.13 Wage Growth by Occupation in Bulgaria
and Poland, 1996–2006 37
x Contents
Tables
1.1 Unemployment Rates by Educational Attainment and
ECA Subregion, 2006 25
2.1 Time Needed to Find First Job by Level of
Educational Attainment in Serbia and Ukraine 67
2.2 Training Participants as Share of Participants in Active
Labor Market Policies and of Total Unemployed,
EU27 and EU10, 2006 77
2A.1 Duration of Education in ECA Countries, Various Years 80
2A.2 ECA Country Results on International Assessments
since 1995 85
3.1 Information Collected from Tracer Study of Dutch
University Graduates, 2007 97
B4.1 Learning Gaps between Roma and Non-Roma
Students in Serbia, as Measured by PISA 2006
and 2009 Results 118
Contents xiii
xv
xvi Foreword
labor market needs. As with assessment, the countries have been slow
to embrace the governance and accountability reforms that are now
part of the landscape of education systems the world over. These limita-
tions to autonomy and accountability for outcomes have reduced the
energy and the incentives for improvements from within the system,
contributing to a shortage of skills. Indeed, firm complaints about the
shortage of relevant skills for expansion and growth have risen to a cre-
scendo in most countries. Finally, the systems increasingly allocate
resources where they are not needed. For example, the sharp decline in
student numbers in the past 20 years has not resulted in a commensu-
rate decline in the number of classrooms and teachers. As a result,
resources are increasingly tied up in buildings and teachers where they
may be better spent elsewhere. Most of these limitations are found not
just in schools but extend to higher education and to training.
The result of these limitations is that the quality and relevance of
education in the countries in the region is increasingly questionable. A
large proportion of students finish lower secondary school with a mini-
mal command of literacy and numeracy. Their failure to learn even the
basics is not picked up early enough by assessment systems, nor are
schools and municipalities given sufficient incentives to ensure that all
students learn the basics. The systems are also struggling with imparting
higher order skills beyond the basics, for which well-motivated and high-
quality teachers are critically needed but which few school systems are
able to attract, given the low pay and historical overstaffing. And without
the vision or resources to make vocational and technical education an
attractive option for students—indeed, this remains one of the most
unreformed subsectors of the education system—quality has eroded and
students have shied away from this form of education, possibly contrib-
uting to the widespread shortage of vocational and technical skills now
found in these countries.
This book makes the case that improving the quality and relevance of
education requires a fundamental change of approach to education in the
countries of the region. To start with, education systems need to “turn the
lights on” and take seriously the measurement of what students actually
learn as opposed to measurement of the inputs into the education process
on the implicit assumption that learning follows. This assessment needs
to inform both teaching and policy making. Policy makers also need to
move away from controlling inputs and processes and instead increase the
emphasis on incentives to improve student learning, whether in school or
in higher education. And, finally, for these reforms to be financially fea-
Foreword xvii
Philippe Le Houerou
Vice President
Europe and Central Asia Region
World Bank
Washington, D.C.
Acknowledgments
and helping to shape the book’s main messages, and Algerlynn Gill for
her excellent research assistance and inputs into several of the book’s
chapters. In addition, Nadezhda Lepeshko and Carmen Laurente helped
prepare graphs and figures for the document. The book also benefitted
from the excellent editorial work of Peggy McInerny and Patricia Carley.
Larry Forgy provided editorial assistance on chapter 1.
We also benefitted from valuable comments from our peer reviewers:
Richard Murnane, Amit Dar, Bernard Hugonnier, Halsey Rogers, and
Manfred Wallenborn. And we benefitted from inputs, comments, and
suggestions from numerous colleagues, including Mohamed Ihsan
Ajwad, Gordon Betcherman, Mary Canning, Isak Froumin, Sachiko
Kataoka, Arvo Kuddo, Toby Linden, Lily Mulatu, Bojana Naceva,
Reehana Rifat Raza, Alberto Rodriguez, Marcelo Selowski, Jan Sadlak,
Jamil Salmi, Luis Crouch, and countless others. Any and all errors that
remain in this volume are the sole responsibility of the authors.
Abbreviations
xxi
xxii Abbreviations
MLT Malta
NLD the Netherlands
NOR Norway
POL Poland
PRT Portugal
ROU Romania
RUS Russian Federation
SRB Serbia
SVK the Slovak Republic
SVN Slovenia
SWE Sweden
TJK Tajikistan
TKM Turkmenistan
TUR Turkey
UKR Ukraine
UZB Uzbekistan
The countries of Europe and Central Asia (ECA)1 are currently emerging
from the deepest recession suffered by any developing region. Post-crisis
conditions are very different from those of preceding years. Financial
resources are more limited and more expensive, and export growth is
restrained by potentially slower growth in destination countries. Restoring
and sustaining growth in this context require reforms that boost com-
petitiveness and increase labor productivity. Such reforms are all the
more important given the shrinking of the working-age population in
many countries of the region.
Earlier boom years in the ECA region exposed significant bottlenecks
to growth, particularly with respect to the skills of the labor force (Mitra
et al. 2010). Paradoxically, for a region with relatively high and expanding
educational attainment (as measured by the number of years of com-
pleted schooling) and relatively high-quality education in the early years
of schooling, a shortage of worker skills has emerged as one of the most
important constraints to firm expansion. This book seeks to answer the
following questions: Why do ECA firms increasingly complain that they
cannot find graduates with the right skills? What can ECA countries do
to close the skills gap?
1
2 Skills, Not Just Diplomas
550
POL EST
2009 mean performance
500 HUN
LVA SVN
TUR HRV CZE
LTU SVK
450 RUS
SRB
BGR
ROM
MNE
400 KAZ
ALB
AZE
350
KGZ
300
0 20,000 40,000 60,000 80,000 100,000
2008 GDP per capita, PPP (current international $)
Source: PISA 2009 Database and World Bank staff calculations based on World Development Indicators Database.
Note: PISA = Programme for International Student Assessment. The figure shows a regression line representing
countries’ predicted PISA reading scores based solely on GDP per capita, compared to the Organisation for Eco-
nomic Co-operation and Development mean reading score (horizontal line) and GDP per capita in 2008. See
“Abbreviations” for a key to country abbreviations.
4 Skills, Not Just Diplomas
Figure O.2 Distribution of Firms in ECA Region that Consider Worker Skills a
“Major” or “Very Severe” Constraint, 2008
9
mean = 30.2
8
TJK
7
UZB
number of countries
6
AZE CZE
5 ARM
%
10
20
30
40
50
60
70
80
90
x<
x<
x<
x<
x<
x<
x<
x<
x<
≤
≤
%
%
10
20
30
40
50
60
70
80
EU10+1 and Turkey Western Balkans LI CIS MI CIS
29
27
student-teacher ratio (primary)
25
23
21
19
17
15
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20
20
20
East Asia and Pacific Europe and Central Asia
Latin America and the Caribbean Middle East and North Africa
Sources: EdStats database, plus authors’ linear interpolations for missing years.
Note: Actual data are marked with dots. One of the problems with international data on student-teacher ratios
(in both the World Bank’s EdStats database and the other international education databases) is that it is unclear
whether or not data for a particular country are reported on a full-time equivalent basis.
has fully escaped the legacy of central planning.4 That system focused
on controlling inputs—that is, it checked whether local actors were in
compliance with detailed norms for all inputs. Education ministries
remained uninformed, however, about whether students acquired skills
and competencies.
In terms of how these impediments affect the different levels of educa-
tion, they are most clearly apparent at the pre-university level, which
absorbs two-thirds of total education funding in the ECA region. The
same three impediments affect tertiary education in a slightly different
way, as this sector has already undergone significant reform over the past
20 years, during which time it has doubled or tripled in size in many
countries (see chapter 5). Because most ECA countries do not yet have
adult education sectors in the modern sense, these impediments cannot
be fairly characterized as problems of this sector. However, the policy
recommendations offered in this book do apply to adult education and
offer ideas and principles for how this subsector could be developed and
managed in the region (see chapter 6).
Overview 9
Figure O.4 Status of Measuring and Using Data on Student Learning Outcomes in
the ECA Region, 2009
several years of
experience in
several years of measuring,
experience with analyzing, and
own assessments making use of
and regular learning results to
early piloting of participation in improve education
own assessment international policy
instruments, assessments
some
participation in Bulgaria
not started, very international
early stages, no, or assessments Hungary
rare participation Latvia
in international Lithuania
assessments Albania
Romania
Croatia
Armenia Serbia
Czech Republic
Azerbaijan Slovenia
Estonia
Belarus
Georgia
Bosnia and
Herzegovina Macedonia, FYR
Kazakhstan Moldova
Kosovo Montenegro
Kyrgyz Republic Poland
Tajikistan Russian Federation
Turkey Slovak Republic
Turkmenistan Ukraine
Uzbekistan
Sources: Authors’ assessments based on data from UNICEF (2007, table 2.1); the extent of countries’ participation
in PISA, Progress in International Literacy Study (PIRLS) and Trends in International Mathematics and Science
Study (TIMSS) since 1995; and inputs from World Bank country experts. The UNICEF data is drawn from “Table 2.1:
Status of Reforms of Assessment and Examination Systems, 2006,” regarding “Introduction of other school exams
or assessments (e.g. basic school)” and “Introduction of sample-based national assessment.” In this table, UNICEF
scores the progress of countries on a range of 0 to 4, with 0 representing “not planned or started” and 4 repre-
senting “operational.” In addition, the figure uses World Bank staff compilations on the number of international
assessments in which each country has participated.
Conclusion
It will take time to steer education systems in the ECA region away from
a focus on inputs toward a focus on delivering a quality education to the
Overview 15
majority of students. Waiting to begin this process will only cause coun-
tries to fall farther behind in delivering the skills their economies need to
compete in a global economy. Overall, much greater measurement of
student learning outcomes is needed at all levels of education if policy
makers are to understand the strengths and weaknesses of their respective
systems and design effective policies to improve them. If the role of cen-
tral governments is redefined to focus on strategic policy, system goals,
and regulation, central policy makers will be freed to use incentives to
manage lower-level actors to meet goals and then measure the results.
In schools, this process will involve greater use of per capita financing,
more autonomy, and greater accountability for learning outcomes. At the
level of higher education, strengthening national quality assurance mech-
anisms is needed to improve the accountability of education providers,
as is greater dissemination of information on learning and employment
outcomes (e.g., results from tracer studies, or university rankings) to
inform student choice. In addition, existing financing incentives will need
strengthening through the introduction of more performance-based bud-
geting and, where warranted, greater institutional autonomy. Although
adult learning systems are critically important for the region, their priority
in individual ECA countries will depend on economic and demographic
conditions and other, perhaps more pressing education challenges. All
countries in the region can, however, begin to develop a coherent strategy
for adult education and training.
Notes
1. The countries included in the Europe and Central Asia (ECA) region include
Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria,
Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakhstan, Kosovo,
Kyrgyz Republic, Latvia, Lithuania, the former Yugoslav Republic of
Macedonia, Moldova, Montenegro, Poland, Romania, the Russian Federation,
Serbia, the Slovak Republic, Slovenia, Tajikistan, Turkey, Turkmenistan,
Ukraine, and Uzbekistan.
2. Adjusted for Purchasing Power Parity, or PPP.
3. The EBRD–World Bank Business Environment and Enterprise Performance
Surveys (BEEPS) provide successive snapshots of the self-reported constraints
to expansion by firms in the ECA region. The surveys were conducted in
virtually all ECA countries in 1999, 2002, 2005, and 2008. Survey samples
are constructed by random sampling from the national registry of firms (or
equivalent) and cover both industry and service sectors.
16 Skills, Not Just Diplomas
4. It should be noted that Turkey does not have a socialist past, but does operate
“in the dark” to a large extent and has not made the performance-oriented
management reforms of other OECD countries. The expenditure efficiency
issues are, however, different in nature. See World Bank 2005a.
5. While the use of learning assessments is not uncontroversial, testing method-
ologies have improved in recent years, making these tests better and more
useful for assessing student learning than was previously the case.
CHAPTER 1
Throughout the world there has been an increase in the demand for
skilled labor in what is now a global knowledge economy. More specifi-
cally, the demand for skills related to complex communication and non-
routine cognitive tasks has increased, while the demand for routine and
manual skills has declined. Employment trends in the region encompass-
ing Eastern Europe and Central Asia (ECA) reflect this new reality: jobs
have been shifting significantly from unskilled to skilled labor and the
wages of highly skilled workers have dramatically improved over the past
20 years. This shift parallels an economic transition that has involved
intensive enterprise restructuring as countries moved from centrally
planned to market-based economies. New skills have come into demand,
as employment has been allocated away from declining industries and
firms toward expanding ones. In addition, there has been a large-scale
shift of jobs from agriculture and, to a lesser extent, industry towards the
service sector.
This evolution in the labor market is at the heart of major changes in
employment outlook throughout the region. The movement toward
greater use of skilled labor has in turn produced relatively greater unem-
ployment for low-skilled workers. Wages of highly skilled white collar
workers have also grown at a much faster rate than those of less-skilled
17
18 Skills, Not Just Diplomas
Figure 1.1 Trends in Routine and Nonroutine Tasks of U.S. Labor Force,
1960–2002
mean task input in percentiles of 1960 task distribution
65
60
55
50
45
40
1960 1970 1980 1990 2000
nonroutine cognitive (analytical) nonroutine cognitive (managerial)
nonroutine manual routine cognitive
routine manual
Source: Autor, Levy, and Murnane 2003. With data updated to 2002 by David Autor.
50
40
30
20
10
0
le en
th new
ta ng
s
w ge
k
as
ob se
of an
ex ki
s
gs
s
k
m
sk
or
st
pl rta
pr ore
ng
ds ch
in
ou
m e
ni
ho or
f
co und
un
on
ar
et e
le
ot
m oos
g
in
on
lv
ch
m
so
Figure 1.3 Job Creation and Destruction Rates for Selected Occupations in
Georgia, 2007
40
35
30
25
percent
20
15
10
5
0
s
ls
bl rs
ry
s
er
an
rk
le
pl
na
m to
ta
sa
cle
ag
s
ici
se era
en
io
er
pe
d
an
hn
ss
em
an
as p
ts
fe
m
d eo
af
te
el
o
ice
cr
pr
an hin
rv
se
ac
m
Source: Data obtained during a labor market survey of Georgian firms in 2007. The survey is discussed in greater
detail in Rutkowski 2008b.
The Demand for Skills in ECA 23
destroyed and many were created for service workers and salespeople,
who thus saw the demand for their skills rise.
The pattern of job reallocation was somewhat different in Ukraine
(see figure 1.4). In 2007, demand shifted away from unskilled towards
skilled manual labor. The job destruction rate exceeded the job creation
rate for elementary occupations;4 consequently, the share of jobs requir-
ing elementary skills decreased. At the same time, the job creation rate
exceeded the job destruction rate for skilled manual occupations, mean-
ing the share of skilled manual jobs increased. There was also a decline in
professional employment in Ukraine.
Figure 1.5 shows patterns of change in the occupational structure of
employment across the ECA region over a longer time series, patterns
that are especially well illustrated by the panels for Poland (panel a) and
Russian Federation (panel d). Within the broad trend of the shift from
unskilled to skilled labor, there are numerous country-specific exceptions.
A diverse group of countries, from Romania to Turkey to the Kyrgyz
Republic, saw increases in the demand for manual labor (not decreases),
but these increases were for skilled manual labor. Some countries (e.g.,
the former Yugoslav Republic of Macedonia) saw a sharp increase in the
Figure 1.4 Job Creation and Destruction Rates for Selected Occupations in
Ukraine, 2007
45
40
35
30
percent
25
20
15
10
5
0
na d
ns
le
bl rs
s
rk
le
er
sio an
m to
op
cia
sa
cle
ls
r
s
bo
se era
er
es rs
pe
ni
d
of age
la
an
ch
as p
ts
d eo
af
te
ice
an
cr
an hin
m
rv
pr
se
ac
m
Sources: Ukraine Labor Demand Survey (World Bank 2009); Bank staff calculations.
24 Skills, Not Just Diplomas
percent
0 –5
–2
–10
–4
–6 –15
–8 –20
ag nd
ns
d ke s
sk rke al
sk rk ual
ke l
ag nd
ns
d ke s
sk rke ral
sk rker al
ke l
or ua
or ua
rk
rk
o u
cia
cia
w ultu
w ultu
an s a
an s a
cle
cle
te ers
ag rs
ille rs
d s
rs
te rs
ag rs
ille rs
rs
un wo n
w an
w an
w an
ille er
a
e
ni
ni
m nal
m nal
d
d
m
m
ric
ric
an
an
ch
ch
o
o
d
d
sio
sio
ille
ice
ice
es
es
rv
rv
of
of
un
se
se
pr
pr
sk
sk
c. Macedonia FYR, 2001–06 d. Russian Federation, 1995–2006
6 4
4 3
2 2
1
percent
percent
0
0
–2
–1
–4
–2
–6 –3
–8 –4
ag nd
ns
ke s
sk rke ral
sk rk ual
ke l
ag nd
ns
d ke s
sk rke ral
ille er l
ke l
or ua
sk ork ua
or ua
rk
or ale
rk
e
ille o al
cia
cia
w ultu
w ultu
an s a
an s a
cle
cle
te ers
rs
ille rs
d s
rs
te ers
ag rs
ille rs
d s
rs
un wo an
w an
un w an
w an
ille er
s
s
ni
ni
m nal
m nal
d
d
m
m
ric
ric
an
an
r
ch
ch
o
o
d
d
sio
sio
ag
w
w
ice
ice
es
es
d
rv
rv
ille
of
of
se
se
pr
pr
sk
sk
0
–2
–5
–4
–6
–8 –10
–10
–12 –15
ag nd
ns
d rke es
ke l
sk rke al
ke l
ag nd
ns
d ker s
sk rke al
ke l
or ra
or ua
or ra
or ua
rk
rk
ille or sale
un wo anu
o u
ille wo sal
cia
cia
w ultu
w ultu
an s a
an s a
cle
cle
te rs
ag rs
ille rs
ille rs
rs
te ers
ille ers
ille rs
rs
w an
w n
w an
a
e
ni
ni
m nal
m nal
d
w d
m
m
k
ric
ric
an
an
ch
ch
d
d
io
io
ag
ice
ice
ss
ss
fe
fe
sk
sk
rv
rv
o
un
se
se
pr
pr
sk
sk
Sources: UNECE SD, Labour Force and Wages; Eurostat SD, Labour Market; World Bank staff calculations.
Note: The arrows in each panel suggest hypothetical movements of workers between occupations (i.e., possible
movements of labor between different occupations during this time period).
The high demand for professional and technical skills in the ECA
region is also evident when one looks at the movement of workers from
unemployment to jobs. Professionals and technicians across countries in
the region have by far the highest chances of escaping unemployment
and finding a new job. Figure 1.6 depicts outflow rates from unemploy-
ment to employment for two ECA subregions: the EU10+1 (European
Union+1) countries and low-income CIS countries (the pattern prevail-
ing in middle-income CIS and South Eastern Europe (SEE) countries is
similar to that in the EU10+1 group).
In the EU10+1 countries, the escape rate from unemployment is
highest among professionals and technicians, followed by medium-
skilled white- and blue-collar workers. It is lowest among unskilled and
agricultural workers. In low-income CIS countries, the pattern is some-
what different: medium-skilled, blue-collar agricultural workers have
better chances of finding a job than do medium-skilled white-collar
workers. But workers in more skilled occupations nevertheless find jobs
more easily than those in less-skilled occupations.
The poor employment prospects of less-skilled workers in the region
are also evident from the unemployment-to-vacancy (U/V) ratio, which is
high for less-skilled workers and low for highly skilled workers. Data on
the U/V ratio for Croatia, the Czech Republic, and Ukraine illustrate the
point (see figure 1.7). Job opportunities for different occupations are quite
similar in these three countries: the number of unemployed professionals
and technicians (skill levels 3 and 4, respectively, in the ISCO-88 frame-
work) per vacancy is low and that of unemployed unskilled workers (skill
level 1) is high. Employment chances are also relatively good for skilled
manual workers, especially craft workers (skill level 2). But job prospects
are limited for white-collar workers with medium-level skills (e.g., clerks,
service workers, salespeople). The demand for white-collar labor with
medium-level skills is limited relative to supply, while the demand for
blue-collar labor with mid-level skills is high relative to supply.
Putting together the information on employment presented in the
previous section with the data on unemployment presented in this
section, it is possible to obtain an estimate of shortages and surpluses
in the economies of the ECA region, particularly by comparing the
relative shares of employment and unemployment of different educa-
tional groups. Figure 1.8 shows that in countries as diverse as Croatia,
Poland, Russia, Uzbekistan, and Tajikistan (listed in order of gross
national income [GNI] per capita), there is a shortage of workers with
professional or technical qualifications and a surplus of workers with
The Demand for Skills in ECA 27
Figure 1.6 Yearly Outflow Rates from Employment to Jobs by Occupation in Two
Subregions of ECA, 2006
a. EU10 + 1
elementary labor
machine operators and
assemblers
craftspeople
agricultural workers
clerks
technicians
professionals
60 70 80 90 100
percent
b. Low-income CIS
elementary labor
machine operators and
assemblers
craftspeople
agricultural workers
clerks
technicians
professionals
60 65 70 75 80 85 90 95 100
percent
Sources: EBRD–World Bank 2006; World Bank staff calculations.
Note: The yearly outflow rate from unemployment to jobs is the number of workers who were employed during
the reference week, but unemployed one year earlier expressed as a percentage of all unemployed.
28 Skills, Not Just Diplomas
a. Croatia, 2006
25
20
15
U/V ratio
10
0
managers
professionals
technicians
clerks
sales and
services
agricultural
workers
craftspeople
machine
operators and
assemblers
elementary
labor
- 4 3 2 2 2 2 2 1
ISCO skill level
10
8
U/V ratio
0
managers
professionals
technicians
clerks
sales and
services
agricultural
workers
craftspeople
machine
operators and
assemblers
elementary
labor
- 4 3 2 2 2 2 2 1
ISCO skill level
c. Ukraine, 2007
10
9
8
7
6
U/V ratio
5
4
3
2
1
0
managers
professionals
technicians
clerks
personal
service
Sales
agricultural
workers
craftspeople
machine
operators and
assemblers
elementary
occupations
- 4 3 2 2 2 2 2 2 1
ISCO skill level
Sources: Data provided by the Croatian Employment Service for 2006, Ministry of Labour and Social Affairs of the
Czech Republic for 2006, and Ukraine State Employment Service for 2007; authors’ calculations.
The Demand for Skills in ECA 29
shortage/surplus index
50
shortage/surplus index
40 60
30 40 shortage
20 shortage
20
10
0
0
–20
–10
surplus
–20 –40
surplus
–30 –60
y
in nal
ry
in l
al
ry
l
ra
in na
ica
es
ica
ar
er
a
a
ne
g
tra tio
tra atio
im
rti
rti
rl
hn
g
en
hn
in
yo
ge
te
te
pr
ca
ec
yg
ec
c
ar
vo
vo
ry
yt
yt
ar
im
da
ar
sic
sic
nd
ar
pr
n
nd
nd
ba
ba
co
co
co
co
se
se
se
se
200
100
shortage/surplus index
shortage/surplus index
150
50 shortage
100 shortage
0 50
0
–50
surplus –50 surplus
–100 –100
s
al
ge
al
y
na
ica
es
es
sit
ar
er
er
le
tio
rti
rl
rl
hn
er
en
en
ol
yo
yo
te
iv
ca
ec
lc
yg
yg
un
ar
ar
vo
yt
ca
ar
ar
im
im
ni
ar
nd
nd
pr
pr
ch
nd
co
co
te
co
se
se
se
shortage/surplus index
150 60
shortage
shortage
100 40
20
50
0
0
–20 surplus
surplus
–50 –40
ic
y
na l/
ge
al
na l/
ra
sit
ar
ar
ar
tio ca
tio ica
s
r
ne
ne
lle
ba
im
nd
rti
l
l
er
ca ni
n
co
ge
ge
te
iv
vo ech
ch
pr
co
un
te
ry
y
se
ca
yt
ar
a
ry
er
nd
nd
vo
ar
a
w
nd
nd
co
co
lo
co
co
se
se
se
se
Sources: Crostat 2007; Bulgaria 2007; Poland 2007; Russian Federation 2007; World Bank 2005b, 2003a; authors’
calculations.
Note: The skill shortage or surplus index is calculated as (e/u–1)*100, where e and u are respectively employment
and unemployment shares of a given occupational group.
30 Skills, Not Just Diplomas
0 2 4 6 8 10 12
percent
Sources: Patrinos 2008 (Ukraine); Kapelyushnikov 2008 (Russia) and Yemtsov, Cnobloch, and Mete 2006 (Belarus,
Bulgaria, Georgia, Hungary, Moldova, Poland, Romania, and Tajikistan).
Note: Average return per a year of additional education estimated based on a Mincerian wage regression of
the form: lnw = α + ρs + β0x + β1x2 + γz + ε; where w represents the wage of an individual, s denotes years of
schooling, x denotes labor market experience (years), z is a vector of other individual and job characteristics
(e.g., gender, industry, etc.), and ε is an error term, which inter alia captures unmeasured differences in innate
ability. The parameter ρ represents the return to one year of schooling.
300
250
200
percent
150
100
50
0
1986 1989 1992 1995 1998 2001 2004
primary vocational secondary technical
secondary general college university
Figure 1.11 Wage Premia by Education Level in the Russian Federation and Turkey
100
80
percent
60
40
20
0
y
ge
ge
ica ry
ve l
ve l
le era
le era
sit
hn a
lle
le
ec nd
l)
l)
er
er en
er n
ol
co
w ge
iv
l/t eco
lc
pp g
un
(u ary
(lo ary
ca
m
s
ni
so
nd
nd
ch
co
co
na
te
se
se
tio
ca
vo
b. Turkey, 2006
120
100
80
percent
60
40
20
0
y
ry
y
ica
ra
ar
ar
a
ne
rti
nd
im
hn
ge
te
pr
co
ec
ry
se
yt
a
ar
er
nd
nd
w
co
lo
co
se
se
Sources: Kapelyushnikov 2008 (Russian Federation); Ercan 2008 (Turkey); authors’ calculations.
The Demand for Skills in ECA 35
percent
percent
200 150
150
100
100
50
50
0 0
s
ke d
le
bl rs
d rke d
cr ers l
le
bl rs
na
na
k a
er
an
rk
er
an
rk
or an
ille o an
or ur
m to
m to
op
op
cle
cle
ag
ag
sio
sio
rs
ag rs
s
ici
ici
w ult
se era
se era
er
er
s w ce
sk s w ice
pe
pe
an
an
n
n
es
es
ric
le rvi
ch
ch
as p
as p
ts
ts
le rv
m
m
of
of
d eo
d eo
af
af
sa se
sa se
te
te
pr
pr
cr
an hin
an hin
ac
ac
m
m
c. Russian Federation, 2007 d. Turkey, 2006
300
250 300
200 250
percent
150 200
percent
100 150
50 100
0 50
–50 0
bl rs
bl rs
s
l
ns
ns
d ke d
d ork and
na
na
le
le
s
s
cr ker al
cr ker al
er
er
m to
m to
rk
rk
n
op
op
cia
cia
or ur
or ur
ag
ag
ille o a
sio
sio
s
s
cle
cle
se era
se era
ric s
ric s
er
er
af s
af s
ag r
ag er
sk s w ice
ille w e
w ult
w ult
pe
pe
ni
ni
an
an
es
es
sk les rvic
as p
as p
ch
ch
rv
ts
ts
r
m
m
of
of
d eo
d eo
sa se
sa se
te
te
pr
pr
an hin
an hin
l e
ac
ac
m
Sources: Eurostat, Labor Market Statistics (database) (Bulgaria and Czech Republic); Ercan 2008 (Turkey),
Kapelyushnikov 2008 (Russia); World Bank staff calculations.
manual workers and clerks, about 1.5 times more. Service workers and
salespeople earn only slightly more—some 10 percent—than workers in
elementary occupations. While patterns across most ECA countries are
similar, the relative wage status of skilled manual workers is much better
in Russia than in the EU10 countries. Concomitantly, the relative wage
status of professionals and technicians is worse in Russia. Moreover,
there are no returns to clerical skills in Russia; clerks earn on average the
same wage as laborers. Thus the occupational wage structure in Russia
is biased towards skilled manual labor; as such, it is closer to that prevail-
ing under central planning. A similar pattern also prevails in Ukraine
(World Bank 2009b).15
In addition to the current variance in wages across job categories,
there is evidence that wage rates are diverging over time. In most ECA
countries for which data are available, workers in occupations requiring
highly skilled labor (e.g., managers, professionals, and technicians) saw
their relative wages increase, while workers in occupations requiring
36 Skills, Not Just Diplomas
medium- and lower-level skills (especially manual skills) saw their rela-
tive wages fall. Thus the wage differential between highly skilled white-
collar workers and less skilled white- and blue-collar workers has
substantially widened. Figure 1.13 illustrates this trend in Bulgaria and
Poland.16 The wage status of highly skilled white-collar workers in these
countries has substantially improved, while that of less skilled workers
(both white- and blue-collar) has deteriorated during the transition.
a. Bulgaria
1.6
1.5
1.4
1.3
wage growth
1.2
1.1
1
0.9
0.8
0.7
1996 2006
managers and officials professionals technicians
clerks service and sales craftspeople
machine operators and assemblers elementary occupations
b. Poland
1.3
1.2
1.1
wage growth
0.9
0.8
0.7
1996 2006
managers and officials professionals
technicians clerks
service and sales agricultural workers
craftspeople machine operators and assemblers
elementary occupations
Sources: Eurostat, Labor Market Statistics (database); World Bank staff calculations.
38 Skills, Not Just Diplomas
5
BIH GEO EST
4
KOS KGZ HRV UKR
3
MKD ALB LVA MDA
2
MNE SRB BGR POL LTU KAZ
1
HUN SVN TUR SVK ROM RUS BLR
0
%
%
10
20
30
40
50
60
70
80
90
x<
x<
x<
x<
x<
x<
x<
x<
x<
<
<
<
<
<
<
<
<
%
%
%
%
60
80
10
20
30
40
50
70
5
4
3
2
1
0
5
.4
.3
.2
.1
x
.
0.
0.
0.
0.
0.
–0
≤
–0
–0
–0
–0
x<
5
x<
x<
x<
x<
x<
x<
x<
x<
x<
x<
0.
≤
≤
1
≤
.
–0
4
.5
.4
.3
.2
0.
0.
0.
0.
–0
–0
–0
–0
4
weeks
0
+1
ey
A
SE
CI
CI
EC
rk
10
Tu
m
m
EU
co
co
in
-in
e-
w
dl
lo
id
m
2.5
2.0
1.5
1.0
0.5
0
+1
ey
A
SE
CI
CI
EC
rk
10
Tu
om
m
EU
co
c
in
-in
e-
w
dl
lo
id
m
There are several ways in which the skills mismatch can be related to
the education and training of graduates in ECA countries. First, gradu-
ates may not have the necessary technical skills to do the jobs that
employers have to offer. This is not solely a matter of educational attain-
ment, that is, of completing a certain level of education. It is also a mat-
ter of whether students have the skills needed on the job—the relevant
knowledge, the ability to apply that knowledge, and the know-how to
complete tasks and solve problems—when they graduate. In addition,
the relevance of the education received by graduates is also important.
That is, students may be studying the wrong things. Second, graduates of
education and training systems in the region may lack the necessary
behavioral (or soft) skills needed by employers, such as job attitudes and
teamwork skills. There is widespread evidence from across the ECA
region that firms seek graduates who not only have knowledge as well
as technical and general skills, but who also have behavioral skills.
For instance, recent surveys of employers in Kazakhstan (panel a of
figure 1.16) and Poland (panel b of figure 1.16) show that firms value
such behavioral skills as highly (in some instances, more highly) as
knowledge and routine cognitive skills (e.g., mathematics and computer
skills)—important as the latter may be.21
Before turning to the ways in which education and training systems
fall short of delivering the right skills, it is worth noting that one possible
reason for the continued identification of skills as a bottleneck to growth
in the ECA region is unreasonable expectations on the part of ECA firms.
A recent comparison of firms’ expectations of new workers in the United
Kingdom and Poland, for example, showed that Polish employers
expected young workers to possess advanced vocational job-specific
skills, while their U.K. counterparts preferred workers who had reached
an adequate level of literacy and numeracy, without necessarily having
fully developed, job-specific vocational skills (World Bank 2010b). The
report concluded that: “The expectations of Polish employers will prob-
ably have to change, and they will have to assume a greater role in the
financing and provision of vocational training to their workforce” (World
Bank 2010b).
Summary
The demand for highly skilled labor has increased in most if not all ECA
countries. This profound change in demand stems from the opening
of their economies to international competition and the consequent
% of firms reporting % of firms reporting
42
0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
advanced
literacy
technical/vocational
and Poland
basic Russian language
technical/vocational
problem solving
problem solving
Skills, Not Just Diplomas
innovative firms
computer skills
(percentage of firms reporting as “very important”)
responsibility
and reliability
b. Poland, 2009
Kazakh language
motivation and
a. Kazakhstan, 2008
commitment
English language
team-work ability
traditional firms
ability to work
independently
communication
soft skills
initiative and time management
ability to work
independently communication
planning and
soft skills
organizing telephone
conversation
customer care
management
skills
Figure 1.16 Employers’ Valuation of Workers’ Knowledge and Skills in Kazakhstan
Firms were categorized as “innovative” if in the year preceding the survey they introduced innovations, otherwise
The Demand for Skills in ECA 43
far-reaching reallocation of labor that took place. New firms have entered
the market, new economic activities have emerged, and new jobs have
been created, calling for different skills than those required previously. As
economies have shifted, the service sector has expanded, while the agri-
cultural and, to a lesser degree, industrial sectors have shrunk. Firms have
changed the skills composition of their workforces to raise productivity
and improve competitiveness; as a result, labor demand has shifted away
from lower-skilled toward higher-skilled occupations.
This change in employment structure is most pronounced in the
countries of Central Europe, but is visible in some form everywhere in
the ECA region. Simultaneously, returns to education have increased and
the wages of highly skilled professional workers have greatly improved,
while those of lesser-skilled workers have deteriorated. Consequently,
the job prospects of better-educated, highly skilled workers have
expanded dramatically. Unemployment and low-paid employment in
the region is accordingly concentrated among less educated and less
skilled workers, a group whose job opportunities are increasingly limited
in most ECA countries.
All of these changes have put added strains on education and training
systems in the region, which are expected to prepare workers for changed
and changing labor markets, but which do not appear geared to meet
those labor market needs. The next chapter looks at the supply of skills
to the labor market in ECA countries in more detail. Given that skills
shortages have become a major constraint to firm growth in many ECA
countries, whatever changes have occurred on the supply side appear to
have been insufficient. The next chapter tries to explain why.
Notes
1. Alternatively, upgrading the education of the workforce may itself be a
factor that leads to increased demand for higher-level skills. That is, the
increase in demand for such skills may reflect an increase in the supply of
these skills, a process known as endogenous skill-biased technological
change. Acemoğlu (1998) argues that a high proportion of skilled workers
in the labor force implies a large market size for skill-complementary tech-
nologies, meaning that the productivity of skilled workers will be upgraded
faster. In this model, an increase in the supply of skills over the long run
induces skill-biased technological change, which shifts the relative demand
for skills (EC 2008).
2. Chapter 2 discusses skills from an education and training perspective.
44 Skills, Not Just Diplomas
education. Yet available evidence indicates that the cohort effect is small
and thus unlikely to account for the differences between the two educa-
tional groups.
9. While not under the scope of this book, it should be noted that migration has
had a significant effect on labor trends in the region. A recent World Bank
study examines both migration and remittances in the region. See Mansoor
and Quillin 2007.
10. Surveys of demand for skills that were carried out in Poland and FYR
Macedonia in 2009 indicated that most newly created jobs require profes-
sional or nonmanual skills, while most jobs that were destroyed required
manual skills. This obviously limits the scope for retraining (Rutkowski 2010;
World Bank 2010b).
11. The low returns to schooling for these countries is presumably in part a legacy
of the socialist period, when education levels were much higher than would
be predicted for their level of development, a situation that holds down wages
during the time when the economic structure is catching up to education
levels.
12. As figure 1.9 suggests, much of the increase in these returns was concentrated
in the early years of the transition. A few countries, typically those with lower
returns, show a flatter trajectory.
13. The terms “premia” and “returns” to education are used here interchangeably.
However, strictly speaking, they are not synonymous. Returns to education
take into account both wage premia and private costs of obtaining education.
Accordingly, premia could have increased without private returns increasing
as dramatically.
14. These estimates do not control for other factors that affect wages, such as
labor market experience, gender, location, and so forth.
15. This finding likely reflects the legacy of central planning and a noncompeti-
tive wage setting, especially in enterprises fully or partly owned by the state,
in both countries.
16. In some countries in the region there were deviations from the overall trend
presented here. For example, salaries of professionals in the Czech Republic
decreased slightly in relative terms between 1996 and 2007. Similarly, in
Estonia and the Slovak Republic, the relative wages of managers fell slightly
during the same period. The slight fall in the relative wages of professionals
and managers that occurred in the mid-2000s likely represents an adjustment
to the increase in the supply of the relevant skills that occurred in response
to growing demand.
17. BEEPS, a joint initiative of the EBRD and the World Bank, collects firm-level
data on a broad range of issues concerning the business environment and the
performance of firms, including labor and workforce skills.
46 Skills, Not Just Diplomas
18. The most common constraint identified by firms in the region was the tax
regime.
19. Firms were categorized as “modern” if they used the Internet in interactions
with their clients and suppliers, and “traditional” if they did not. Clearly, use
of the Internet is only a proxy for modern technology.
20. Some degree of skills mismatch is natural and unavoidable in every growing
and restructuring economy due to adjustment costs and the delayed response
of national educational systems. Especially in the context of fast-changing
technology, changes in the skill structure of the labor supply tend to lag
behind those of labor demand, even in a well-performing labor market.
21. Surveys of the demand for skills in Poland and FYR Macedonia that were car-
ried out in 2009 provide additional empirical support to the claim that
employers value job attitudes and behavioral skills at least as much as aca-
demic or technical and/or vocational skills. At the same time, both Polish and
Macedonian employers complain that many recent school graduates and
young workers are lacking “soft” skills (Rutkowski 2010; World Bank 2010b).
CHAPTER 2
47
48 Skills, Not Just Diplomas
more than having access to a tool and the technical skills required to
handle it (e.g., the skills needed to read a text or use software). Rather,
individuals also need to create and adapt knowledge and skills, which
require a familiarity with the tool itself, as well as an understanding of
how it changes the way one can interact with the world and how it can
be used to accomplish broader goals.
The ability to interact in heterogeneous groups, including the ability to
relate well to others, cooperate and work in teams, and manage and
resolve conflicts. As societies become more fragmented and more
diverse, it becomes important to manage interpersonal relationships
well, which benefits both individuals and their abilities to build new
forms of cooperation.
Box 2.1
• Skills: the ability to apply knowledge and use know-how to complete tasks and
solve problems. Skills may be cognitive (involving the use of logical, intuitive,
and creative thinking) or practical (involving manual dexterity and the use of
methods, materials, tools, and instruments).
• Competence (or competency): the proven ability to use knowledge; skills; and
personal, social, and methodological abilities in work or study situations and in
professional and personal development. Competence may be described in
terms of responsibility and autonomy.
• Learning outcomes: statements of what a learner knows, understands, and is
able to do on completion of a learning process, defined in terms of knowledge,
skills, and competencies.
• Qualifications: a formal assessment and validation by a competency body that
determines an individual has achieved learning outcomes in conformance with
given standards.
• Knowledge: the outcome of the assimilation of information through learning.
Knowledge is the body of facts, principles, theories, and practices related to a
field of work or study. It can be theoretical or factual or both.
Source: EU 2008.
50 Skills, Not Just Diplomas
The ability to act autonomously, including the ability to form and con-
duct life plans and personal projects, as well as to assert rights, interests,
limits, and needs. Acting autonomously does not mean functioning in social
isolation. Instead, individuals must act autonomously in order to partici-
pate effectively in the development of society and function well in differ-
ent spheres of life, including the workplace, family life, and social life.
To develop the higher-level competencies in demand among contem-
porary employers, students must become able to organize and regulate
their own learning, learn both independently and in groups, and over-
come difficulties in the learning process. This versatility requires that they
become aware of their own thinking processes, as well as their learning
strategies and methods. What they acquire from school and university
education is the cognitive and motivational prerequisite for successful
learning in future life, with adult education and training helping refresh
and build on these foundations.
Education and training systems today are assessed against their ability to
develop the above-described competencies in their students. Clearly, such
systems must first provide individuals access to schooling and training, as
measured by years of education of recent graduates, current enrollment
trends, and the share of working-age adults who participate in adult learn-
ing programs. Schooling must, moreover, be of sufficient quality in order for
it to result in actual learning, or the acquisition of knowledge and skills
(sometimes referred to as educational outcomes) by students. Various
national and international student assessment instruments have been devel-
oped precisely to measure these learning outcomes. Finally, an education
system must be able to impart the more complex competencies described
above in graduates at all levels; that is, it must impart education that is
relevant with respect to labor market demands and social conditions.
The most rapid growth in attainment has taken place at the tertiary
level, albeit with considerable variation across the region. In fact, the
increase in tertiary enrollment rates over the period 1999–2006 is higher
for ECA countries (from below 40 percent to almost 50 percent) than for
either high-income countries or the East Asia or Latin America regions.
This growth has occurred across the region over the past decade, although
Turkey and the South Eastern European and low-income CIS countries
are trailing the EU10+1 and middle-income CIS countries. Gross tertiary
enrollment rates in the ECA region as a whole demonstrate a pattern
similar to those in secondary education: a dip in the mid-1990s, followed
by a strong rebound. For the period 1990–2006 overall, the level of gross
tertiary enrollments in the region were second only to the high-income
countries (see figure 2.1).
ECA countries not only do well in terms of attainment, they also
deliver relatively good-quality education, especially in the early grades.
Indeed, many younger students in the region outperform their Western
European peers in reading at the fourth-grade level. For example, stu-
dents from the majority of participating ECA countries scored above
the scale average2 on the Progress in International Reading Literacy
Study (PIRLS) in both 2001 and 2006. There are, however, important
Figure 2.1 Gross Enrollment Rates in Tertiary Education, by World Bank Region
(percent)
80
70
60
50
40
30
20
10
0
lA d
Af and
cif d
ia
be nd
Af aran
e
ag
ra an
Pa an
As
rib a a
sia
ic
an
er
ric
ric
nt pe
rth ast
ia
h
h
Ca ic
av
Sa
ut
As
Ce uro
e er
No e E
So
b-
CD
st
th m
dl
Su
E
Ea
OE
n
id
tin
r
M
te
La
s
Ea
variations across the ECA region. Countries that have not joined the EU,
for example, scored below the average, with the marked exception of the
Russian Federation, which scored well above the average and demon-
strated the greatest improvement between 2001 and 2006 (see fig-
ure 2.2). This strong performance in reading at the fourth-grade level
indicates that schools in ECA countries are very capable of providing a
quality education at the early stages.
Many—but not all—ECA countries continue to do well on interna-
tional assessments of reading at or near the end of lower secondary educa-
tion (e.g., Programme for International Student Assessment [PISA],
Trends in International Mathematics and Science Study [TIMSS]) relative
to the performance of other countries at the same income level.3 Figure
2.3 provides a scatter plot of 2008 per capita GDP against the PISA 2009
reading scores of countries participating in PISA (with ECA countries
highlighted by their letter acronyms). The figure shows a positive rela-
tionship between the two values, suggesting that 15-year-olds in countries
with higher national incomes tend to perform better on the reading com-
ponent of the PISA. Significantly, the ECA region seems divided. Certain
countries (mainly the new EU member states but also Turkey and Serbia)
lie above the regression line, indicating that the reading performance of
their students is better than would be predicted on the basis of GDP per
capita alone. Others lie below the regression line, including Romania,
Montenegro, Bulgaria, Azerbaijan, and the Kyrgyz Republic. Students in
these countries scored lower on the reading component of PISA than
their national per capita GDP would have predicted.
600
550
mean performance
500
450
400
350
300
Macedonia,
FYR
Bulgaria
Czech Republic
Hungary
Latvia
Lithuania
Poland
Romania
Slovak Republic
Slovenia
Georgia
Moldova
Russian
Federation
Turkey
South EU10+1 Low-CIS Mid- Turkey
Eastern CIS
Europe
2001 2006 PIRLS scale average
Source: World Bank staff calculations based on PIRLS test score data downloaded from EdStats Database.
Note: Thirty-five countries from around the world (although with limited participation from Latin American countries and no participation from Sub-Saharan Africa) participated in PIRLS
2001. In 2006, 41 countries participated, 15 of which were Western European countries. (A list of participating countries is available at http://nces.ed.gov/surveys/pirls/countries.asp.)
53
54 Skills, Not Just Diplomas
550
POL EST
2009 mean performance
500 HUN
LVA SVN
TUR HRV CZE
LTU SVK
450 RUS
SRB
BGR
ROM
MNE
400 KAZ
ALB
AZE
350
KGZ
300
0 20,000 40,000 60,000 80,000 100,000
2008 GDP per capita, PPP (current international $)
Sources: PISA 2009 Database and World Bank staff calculations based on World Development Indicators Database.
Note: The figure shows a regression line representing countries’ predicted PISA scores based solely on GDP per
capita, compared to the OECD mean reading score (horizontal line) on the assessment and GDP per capita in
2008. See “Abbreviations” for a key to country abbreviations.
Box 2.2
0
10
20
30
40
50
60
70
80
90
Kyrgyz Republic
(percent)
Azerbaijan
Kazakhstan
Albania
Montenegro
Bulgaria
Romania Section of PISA 2009
Russian Federation
Serbia
Turkey
ECA
Lithuania
Czech Republic
Croatia
Slovak Republic
Slovenia
Latvia
Hungary
Poland
Estonia
Mexico
Chile
Austria
Israel
Source: World Bank Staff calculations based on OECD PISA 2009 database.
Greece
Italy
France
Spain
Germany
United Kingdom
Belgium
Portugal
United States
Sweden
OECD, non-ECA
Ireland
Iceland
Switzerland
Denmark
Norway
New Zealand
Netherlands
Australia
Japan
Canada
Finland
Korea, Rep.
Panama
Peru
Qatar
Indonesia
Argentina
Tunisia
Brazil
Jordan
Trinidad and Tobago
Colombia
Thailand
Figure 2.4 Assessing a Competency—Reading Level at Age 15: Share of Students Scoring Level 1 or Below on Reading
Others
Uruguay
Dubai (UAE)
Liechtenstein
Chinese Taipei
Macao SAR, China
Singapore
Hong Kong SAR, China
Shanghai, China
Education and the Supply of Skills to the ECA Market 57
Box 2.3
800 45
700 679 40
600 35
528
500 435 30
percent
536
score
25
400 382 20
387
300 15
200 10
100 5
0 0
s
hi e
ol l
s
ho a
ol
ol
ol
ol
ac siv
sc tion
ho
ho
ho
ho
ng
s
te n
e te
sc
sc
sc
sc
gh ca
ag h in
hi vo
gh
gh
h
ig
ig
hi
hi
nu it
lh
lh
ia
la w
e
na
ra
ol
nc
ia
n ol
ne
at
ol
tio
ig o
ie
re ch
An
at
ge
sc
ca
An
fo h s
vo
g
hi
in the other two types of high schools belong to the richest 20 percent of house-
holds in the country. These asymmetries in access to different types of public
schools, and the variable quality of education that these schools provide, consti-
tute a formidable barrier to improving the quality of education for all in Turkey.
International evidence shows that very selective systems that track students at an
early age, such as the Turkish system, not only significantly increase educational
inequality, but also reduce a country’s average educational performance
(Hanushek and Wößmann 2008).
On a positive note, Turkish students did significantly better on the PISA 2009
test (compared to their performance in 2006) in all tested areas: science, reading,
and math. This improvement may be a first indication that recent reforms, most
notably the 2004 reform in the primary school curricula—which moved to a stu-
dent-centered, constructivist approach—may have started to pay off. Despite the
improvements, though, a significant proportion of 15-year-olds (25 percent in
reading, 30 percent in math, and 42 percent in science) in Turkey continue to
perform below even the most basic proficiency levels. It is critical to remember,
moreover, that the PISA results represent only the 57 percent of 15-year-olds who
are still in the educational system.
Source: PISA 2009 Database; World Bank forthcoming.
Box 2.4
same level of proficiency from tertiary education as did the smaller, more
qualified university cohorts of the past.
In addition, a great deal of the expansion of tertiary education has
occurred in new institutions whose short-term motives may be driven more
by profits than a need to guard their institutional reputations. Although
private, profit-based tertiary education providers can create a more flexible,
responsive university system (see chapter 5), there are signs of persistent
problems in this sector, including mounting quantitative (as well as anec-
dotal) evidence of fraud, corruption, and unethical behavior—the scope of
which is extensive—raising doubts about its capacity to ensure that stu-
dents graduate with the requisite skills (and not just diplomas). For example,
a recent survey of university students in several ECA countries found that
more than 60 percent of respondents knew other students who had pur-
chased either entrance to the university or a specific grade (see figure 2.5).
Finally, a large part of the expansion of tertiary education took place
in part-time (or weekend) and long-distance learning programs. The
format of these programs was new and their quality unknown. In 2009,
for example, half of Poland’s undergraduate students were enrolled in
weekend programs. In Romania in the same year, 42 percent of students
were enrolled in private universities, 66 percent of whom were part-time
or long-distance students. It is impossible to determine whether these
Figure 2.5 Students Aware of Bribery for Grade or Exam in their Faculty
90
80
70
60
percent
50
40
30
20
10
0
Bulgaria Croatia Moldova Serbia Kyrgyz
Republica
students are receiving the same quality education as their peers in full-
time, regular programs.
Box 2.5
70
enrolled in vocational programs
60
50
40
30
20
10
0
A
Af and
cif d
tin Asia d
er tral
Af aran
pe
Pa an
t n
EC
ro
es a
a
ic
ica
ica
Am en
ric
rth ast
ia
W th
Eu
r
Sa
C
As
u
No E
rn
d
So
b-
e
st
an
te
dl
Su
Ea
es
id
W
M
La
0
10
20
30
40
50
60
70
80
90
100
(percent)
Source: UIS.
Hungary
Lithuania
Estonia
Latvia
Poland
EU10+1
Slovenia
Romania
Croatia
Slovak Republic
Czech Republic
1999
Albania
Macedonia, FYR
SEE
1989 Serbia
Belarus
Kazakhstan
2007
MI CIS
Ukraine
Russian Federation
Azerbaijan
Armenia
Figure 2.6 Share of Upper Secondary Students Enrolled in Vocational Programs in ECA Countries over Time
Georgia
Tajikistan
LI CIS
Kyrgyz Republic
Moldova
Uzbekistan
Turkey
Turkey
Education and the Supply of Skills to the ECA Market 67
Table 2.1 Time Needed to Find First Job by Level of Educational Attainment
in Serbia and Ukraine
(percent)
a. Serbia
Secondary Secondary Secondary
Primary trade vocational general College University
No search 7.0 5.7 4.7 9.6 7.7 11.7
1–3 months 22.0 25.7 22.3 19.2 23.1 26.0
4–6 months 8.4 7.5 7.2 6.7 9.1 12.8
7–12 months 10.6 10.8 12.4 8.7 18.0 9.4
1–2 yrs 15.0 15.2 19.5 8.4 13.2 21.6
> 2 yrs 37.0 35.1 33.9 47.3 29.0 18.4
Total 100 100 100 100 100 100
Source: Huitfeldt, Johansen, and Kogan 2008.
b. Ukraine
Less than Secondary Secondary Junior
secondary vocational general specialist University
No search 0.9 7.3 4.8 9.0 26.6
1–3 months 32.7 49.5 31.4 58.2 44.4
4–6 months 6.7 7.6 10.5 7.4 8.4
7–12 months 7.5 6.9 9.6 6.1 5.2
1–2 yrs 5.2 11.8 9.1 4.5 5.4
> 2 yrs 47.0 16.9 34.7 14.7 10.0
Total 100 100 100 100 100
Source: Huitfeldt, Johansen, and Kogan 2008.
In fact, many countries in the ECA region have not yet begun to ade-
quately promote adult learning as a means of addressing the current skills
deficit and improving worker productivity in the face of declining demo-
graphics. Although a substantial share of workers participate in adult
education and training programs in several ECA countries, the percent-
ages vary across the region. Even in countries that have relatively high
availability of continuing vocational education and training programs, the
people who take part in these programs tend to be already skilled workers,
rather than nonskilled and “nonproductive” workers. Finally, it seems that
in many ECA countries, programs to help retrain or reskill the unem-
ployed only reach a few of the unemployed.
Adult learning programs (see box 2.6) have remained a blind spot in
education and training policy across much of the ECA region. A key
component of the socialist production process, worker training has under-
gone substantial changes as a result of the transition to market economies.
Under communism, such training focused exclusively on continuing
vocational training of the employed workforce. With the exception of
socialist Yugoslavia, unemployment in these countries was nonexistent by
definition, hence adult education and training of the unemployed was not
a feature of their education and training systems.
Partly, adult education and training remains a blind spot in the region
because remarkably few data on the sector are available. While several
OECD countries have been conducting national surveys of adult train-
ing in firms for many years, such surveys have not been comparative
across countries and data for ECA countries is sparse. Even OECD
countries focus on the quantity of this learning (i.e., number of partici-
pants, hours of learning), not its quality and relevance. For the ECA
region as a whole, the 2005 and 2008 Business Environment and
Enterprise Performance Surveys (BEEPS 2005, 2008) included some
limited questions on adult training by firm size and some degree of
worker characteristics; however, problems relating to cross-country
comparability make it difficult to reach solid conclusions. In the EU, the
firm-based Continuing Vocational Training Surveys (CVTS) of 1999
and 2005, which offer information on worker training, and the Adult
Education Survey (AES) of 2007, which offers data on individual par-
ticipation in formal or nonformal education and training, are the first
surveys that permitted cross-country analysis of adult education and
training patterns. Both included a number of ECA countries and serve
as the basis of much of the analysis in this book. The data used here,
therefore, center disproportionately on the EU10+1.
Education and the Supply of Skills to the ECA Market 69
Box 2.6
Continuing Vocational Education and Training (CVET). These programs are aimed
at people who are currently employed. First, they offer in-service training designed
to aid employees in acquiring new competencies or improving existing ones rel-
evant to their firms’ operations. In-service training plays a critical role in increasing
the human capital of the existing workforce, addressing skills depletion, and keep-
ing older workers productive for longer, as well as alleviating skills mismatches.
Second, these programs offer education and training to individuals who seek to
develop skills that will raise their chances of moving to a better job (that is, educa-
tion and training not related to their current job).
Figure 2.7 Share of ECA Firms that Offer Formal Training Programs for Permanent, Full-Time Employees, 2008
80
69 71
70 66
61
60
52
50 46 47 44
43
percent
41
40 37
33 33
31 30 30
30 28 29
25 25 25 25
21
19 20
20
15 15
10 11
10
0
HUN
ROU
BGR
LVA
LTU
SVN
POL
EST
CZE
MKD
ALB
Kosovo
MNE
TUR
SRB
BHR
UKR
KAZ
BLR
RUS
UZB
AZE
GEO
KGZ
ARM
MDA
HRV
SVK
TJK
EU10+1 SEE MI CIS LI CIS
Source: World Bank Staff calculations based on raw dataset for EBRD-World Bank 2009.
Note: LI = low-income, MI = middle-income. See “Acronymvs and Abbreviations” for a key to country abbreviations.
Education and the Supply of Skills to the ECA Market 71
50
40
percent
30
20
10
0
EU27
CZE
SVN
EST
POL
ROU
HUN
BGR
LVA
LTU
LUX
FRA
SWE
BEL
FIN
NLD
ESP
AUT
GBR
MTL
DEU
CYP
ITA
NOR
PRT
SVK
DNK
EU10 other EU
70
60
50
percent
40
30
20
10
0
HUN
POL
LVA
LTU
BGR
EST
GRC
ITA
ESP
FRA
CYP
AUT
GBR
DEU
NOR
FIN
SWE
SVK
EU10 other EU
Source: World Bank staff calculations based on CVTS 2005 (a) and AES 2007 (b).
Note: See “Abbreviations” for a key to country abbreviations and groupings.
to BEEPS 2005 data, skilled workers in all ECA countries are more likely
to participate in education and training than other workers in the same
firm. Figure 2.10 shows that overall, firm-based worker training data
mask substantial differences in training rates between skilled labor on
one hand and nonskilled and so-called nonproductive workers on the
other, with the latter two groups having a lower (sometimes much
lower) participation rate than that of skilled workers. This finding is
Education and the Supply of Skills to the ECA Market 73
40
30
20
10
0
HUN LVA LTU ROU EST BGR POL EU27 SVK SVN CZE
8
6
4
2
0
CZE SVN SVK EU27 EST HUN POL LTU ROU BGR LVA
Figure 2.10 ECA Firms that Offer Formal Training to Employees, by Employee Category, 2005
(percent)
90
80
70
60
percent of firms
50
40
30
20
10
a
B
R
Z
U
O
M
DA
S
R
N
KD
Z
A
T
N
V
K
K
B
PO
BE
AZ
RU
CZ
ES
TU
BG
UK
UZ
AL
KA
TJ
KG
SV
HR
LV
RO
LT
BI
HU
SV
GE
AR
SR
M
M
Figure 2.11 (panel a) presents 2005 BEEPS data that confirm the
internationally well-established fact that large firms provide more training
than small firms. The bottom panel presents data from the 2005 CVTS
on the best- and worst-performing ECA countries on this measure—the
Czech Republic and Bulgaria, respectively—and compares this data with
the best-performing EU country in the CVTS sample, the United
Kingdom. In Bulgaria and the Czech Republic, there is a wide gap in the
share of companies that offer training between small and large enter-
prises, although at different levels, while the United Kingdom shows little
divergence between small and large firms.
Figure 2.11 Large and Small Firms that Offer Training to Skilled Employees,
ECA, EU, and United Kingdom, 2005
(percent)
90
80
70
percentage of firms
60
50
40
30
20
10
0
Ba
O
B
TU E
R
RUR
AR S
M
U
Z
M U
M A
HUD
N
SR R
Z
A
H
L
T
B
L
E
N
SV V
K
K
PO
BE
AZ
CZ
ES
BG
UK
UZ
AL
TJ
KA
KG
SV
HR
D
LV
RO
LT
BI
K
GE
80
70
60
50
40
30
20
10
0
Bulgaria Czech Republic United Kingdom
Sources: (a) World Bank staff calculations based on raw dataset of EBRD-World Bank 2005. (b) World Bank staff
calculations based on CVTS 2005.
Notes: BEEPS defines small firms as fewer than 50 employees, medium firms as between 50 and 249 employees,
and large firms as more than 250 employees. See “Abbreviations” for a key to country abbreviations.
a. Results for Serbia (SRB) include Montenegro.
Education and the Supply of Skills to the ECA Market 77
Summary
This chapter has sought to understand why employers in the ECA region
increasingly complain of a skills shortage despite high levels of educa-
tional attainment, respectable educational quality, and booming tertiary
enrollments in ECA countries. Although the arguments presented here
can necessarily be based only on the data currently available, the findings
nevertheless point to problems with the limited quality and relevance of
education as contributing sources of the skills deficit. Graduates of upper
secondary and tertiary education appear to either not be learning enough
while in school or acquiring skills that are not in demand on the labor
market. In addition, the decline of vocational schools at the upper sec-
ondary level may have intensified the skills shortage.
78 Skills, Not Just Diplomas
Figure 2.12 Adult Education and Training Program Participants in EU, 2007
50
40
30
20
10
0
HUN
POL
LTU
LVA
BGR
EST
GRC
ITA
CYP
ESP
FRA
AUT
DEU
GBR
FIN
NOR
SWE
SVK
EU10 other EU
50
40
30
20
10
0
HUN
POL
LVA
LTU
EST
BGR
GRC
ITA
ESP
FRA
CYP
AUT
DEU
GBR
NOR
FIN
SWE
SVK
EU10 other EU
employed unemployed inactive
86
TIMSS PISA PIRLS
Math grade 8 Math grade 4 Science grade 8 Science Math Science Reading Reading *
Country
Name 1995 1999 2003 2007 2003 2007 1995 1999 2003 2007 2003 2007 2000 2003 2006 2009 2000 2003 2006 2009 2000 2003 2006 2009 2001 2006
Bosnia and
Herzegovina 456 466
Kazakhstan 549 533 405 400 390
Ukraine 462 469 485 474
Georgia 410 438 421 418 471
Tajikistan
Turkmenistan
Uzbekistan
Belarus
Kosovo
Sources: IEA 1995, 1999, 2003, 2007; PISA Databases 2000, 2003, 2006, 2009; IES 2001, 2006.
Note: * Authors’ summary assessment of trends in countries’ scores across time and across various tests.
Education and the Supply of Skills to the ECA Market 87
Notes
1. Along lines similar to the OECD DeSeCo project, the European Qualifications
Framework for Lifelong Learning identifies eight key competences: (1) com-
munication in the mother tongue; (2) communication in foreign languages;
(3) mathematical competence and basic competence in science and technology;
(4) digital competence; (5) learning to learn; (6) interpersonal, intercultural,
and social competences, together with civic competence (7) entrepreneurship;
and (8) cultural expression. In addition, critical thinking, creativity, initiative
taking, problem solving, risk assessment, decision making, and managing
feelings constructively are seen as playing a role in all eight key competencies
(EU 2008). The EU framework identifies the essential knowledge, skills, and
attitudes related to each.
2. This is a scale metric established by setting the mean scores of participating
countries to 500 and the standard deviation to 100, thus enabling compari-
sons over time, since all cycles are placed on this metric so that scores are
equivalent from cycle to cycle. In contrast, the international average, obtained
by averaging across the mean scores for each of the participating countries,
needs to be recomputed for each new cycle, based on the set of participating
countries and changes from cycle to cycle, depending on the set of countries
taking part.
3. Income level is used as a rough measure of the quality of all nonschool factors,
including parental inputs, that help students learn.
4. Some ECA countries are making rapid progress in addressing this information
gap (see chapter 5).
5. As noted in chapter 1, footnote 6, care should be taken when comparing
employment outcomes of vocational vs. general secondary graduates. In par-
ticular, it can be argued that better employment outcomes of workers with
technical secondary education compared with those of workers with general
secondary education reflect the different composition of both groups. One
hypothesis is that due to the selection process, graduates of general secondary
schools who do not proceed to tertiary education are less able, on average,
than the graduates of secondary technical schools.
6. The higher incidence of training among skilled than among nonskilled work-
ers is consistent with evidence from across ECA and the OECD.
CHAPTER 3
This chapter identifies the major impediments that prevent the educa-
tion system from delivering the skills that are being demanded by the
labor markets as identified in the previous chapters. It then offers a policy
framework and policy directions to help overcome these impediments.
The impediments and a policy framework for addressing them guide the
discussion of education system reform for the remainder of the book.
Here we argue that the three most pressing problems of ECA educa-
tion systems that get in the way of imparting the needed skills are:
(1) the lack of data on the skills and competencies that students actually
acquire as a result of the educational process; (2) the legacy of central
planning—particularly its effect on the management of education sys-
tems in the region—which makes education systems inflexible and resis-
tant to the initiative of front-line actors (who can actually improve
education); and (3) the inefficient use of resources, a problem that has
become particularly acute in the current economic climate. Certainly
these are not the only impediments to making education systems more
responsive to skills needs in the region. However, they are common to all
ECA countries and lie at the heart of addressing the skills deficiencies
discussed in chapter 2.1
89
90 Skills, Not Just Diplomas
100
6
90 17
25 26
80 40 31
70
60
56
50
61 61
40
30 57 63
20
27
10
14 12
0 3
EU10+1 MI CIS W. Balkans LI CIS Turkey
compulsory or less upper secondary beyond upper secondary
and training. At best, current surveys in that sector measure the number
of training hours and courses in which individuals participate, or whether
a firm offers training. But no international comparative data yet exist to
compare the quality or relevance of such training.
Box 3.1
assessing
learning
outcomes using
analysis to
identifying improve
employment analyzing and
disseminating policies
outcomes of
graduates data
e.g.,
performance-
collecting
based
data
budgeting
Source: Authors.
Figure 3.3 Status of Measuring and Using Data on Student Learning Outcomes in
the ECA Region, 2009
several years of
experience in
several years of measuring,
experience with analyzing, and
own assessments making use of
and regular learning results to
early piloting of participation in improve education
own assessment international policy
instruments, assessments
some
participation in Bulgaria
not started, very international
Hungary
early stages, no, or assessments
rare participation Latvia
in international Lithuania
Albania
assessments Romania
Croatia
Armenia Serbia
Czech Republic
Azerbaijan Slovenia
Estonia
Belarus Georgia
Bosnia and Macedonia, FYR
Herzagovina Moldova
Kazakhstan Montenegro
Kosovo Poland
Kyrgyz Republic Russian Federation
Tajikistan Slovak Republic
Turkey Ukraine
Turkmenistan
Uzbekistan
Sources: Authors’ assessments based on data from UNICEF (2007, table 2.1); the extent of countries’ participation
in PISA, PIRLS, and TIMSS since 1995; and inputs from World Bank country experts. The UNICEF data is drawn from
“Table 2.1: Status of Reforms of Assessment and Examination Systems, 2006,” regarding “Introduction of other
school exams or assessments (e.g. basic school)” and “Introduction of sample-based national assessment.” In this
table, UNICEF scores the progress of countries on a range of 0 to 4, with 0 representing “not planned or started”
and 4 representing “operational.” In addition, the figure above uses World Bank staff compilations on the number
of international assessments in which each country has participated.
Note: As a signal that data-driven policy is a way to build capacity, the year in which countries first participated in
an international assessment was one of the indicators used to group countries.
between practical and theoretical knowledge?). Table 3.1 shows the types
of information that the survey collects.
Box 3.2
• 1 deputy school director for 11–24 class groups; 2 if more than 38 class groups
• 1 managing deputy director if an urban school has more than 600 students or
if a rural school has more than 400 students
• 1 pedagogue if a school has more than 8 class groups (regardless of the total
number of students in the school)
• 1 extracurricular activities group leader per each 16 class groups (regardless of
the total number of students in the school)
• 1 cleaner per 500 square meters (0.5 per each 250 square meters)
• 1 coat room attendant per 200 coat spots in school
• 1 yardkeeper per 1.5 hectares of school property
Figure 3.4 Relationship between Primary School Test Scores, Adult Education
Levels, and Per Student Spending in Two Municipalities of Poland
30
Rutka-Tartak
28 Zl 3,710
sixth-grade test score (max = 40)
26
24
22
20
18
16 Tarłów
Zl 8,330
14
10.0 12.5 15.0 17.5 20.0 22.5 25.0 27.5 30.0 32.5 35.0 37.5 40.0 42.5 45.0 47.5 50.0
adults with secondary education or higher, as % of total
Source: Authors’ analysis of data from the Ministry of Finance, Central Statistical Office, Ministry of National Edu-
cation, and Central Examination Board, Poland, and the World Bank’s BOOST database; Kheyfets and others 2011.
Note: Size of bubble indicates relative size of per student spending.
Resolving the Skills Shortage in the ECA Region: A Policy Framework 101
Figure 3.5 Relationship between Primary School Test Scores, Average Class Size,
and Per Student Spending in Two Municipalities of Poland
30
Rutka-Tartak
28 Zl 3,710
sixth-grade test score (max = 40)
26
24
22
20
18
16
Tarłów
14 Zl 8,330
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
average class size
Source: Authors’ analysis of data from the Ministry of Finance, Central Statistical Office, Ministry of National Edu-
cation, and Central Examination Board, Poland, and the World Bank’s BOOST database; Kheyfets and others 2011.
Note: Size of bubble indicates relative size of per student spending. See end note for data sources used for this
analysis.
130
120
110
100
90
80
70
60
90
92
94
96
98
00
02
04
06
19
19
19
19
19
20
20
20
20
29
27
student-teacher ratio (primary)
25
23
21
19
17
15
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20
20
20
Source: EdStats database, plus author’s linear interpolations for missing years.
Note: Actual data are marked with dots. One of the problems with international data on student-teacher ratios
(in both the World Bank’s EdStats database and other international education databases) is that it is unclear
whether or not data for a particular country is reported on a full-time equivalent basis.
106 Skills, Not Just Diplomas
These low ratios are problematic for two reasons. First, other things
being equal, lower student-teacher ratios imply that it is more costly to
finance education per student. There is no evidence, however, that these
rising costs have resulted in better-quality education (for instance, see
figure 3.8 for the case of Romania, where TIMSS scores have remained
roughly flat for a decade). Second, lower student-teacher ratios mean that
more of the education sector’s scarce resources are tied up in employee
costs, leaving very few resources available for either innovations or much-
needed classroom learning materials (e.g., new textbooks, computers,
Internet access).
Figure 3.8 Real Per Student Expenditure Compared to TIMSS Math Scores in
Romania, 1999–2008
350
200
TIMSS score
300
150 250
200
100
150
100
50
50
0 0
99
00
01
02
03
04
05
06
07
08
19
20
20
20
20
20
20
20
20
20
Sources: Authors’ calculations based on TIMSS (various years) and expenditure data obtained from Romania’s
Ministry of Public Finance.
Resolving the Skills Shortage in the ECA Region: A Policy Framework 107
60
50
40
percent
30
20
10
0
Bulgaria
Hungary
Lithuania
Romania
Slovak Republic
Slovenia
Macedonia, FYR
Turkey
Russian
Federation
Armenia
EU10+1 SEE MI LI
CIS CIS
countries
1995 1999 2003 2007
Sources: IEA (1995, 1999, 2003, and 2007): Mathematics and Science Teacher Background Data Almanacs.
Note: LI = low-income, MI = middle-income.
per student
financing widely
introduced
piloting of per
student financing
Armenia (2005)
Bulgaria (2008)
per student Kyrgyz Republic
financing under (2006) Czech Republic
discussion Moldova (2010) (1992)
Russian Federation Estonia (2001)
(1998) Georgia (2007)
Azerbaijan
not started or Tajikistan (2005) Hungary (1990)
Latvia
very early Uzbekistan (2008) Kosovo (2002)
stages in Macedonia, FYR
Lithuania (2001)
discussions Serbia
Poland (2000)
Slovenia
Albania Romania (2010)
Belarus Slovak Republic
Bosnia and (2004)
Herzegovina
Croatia
Kazakhstan
Montenegro
Turkey
Ukraine
student” rather than inputs (e.g., staff). Money can follow the student in
many forms, such as student vouchers and block grants tied to student
numbers. Countries that introduce per student financing typically design
formulae that make it explicit and transparent how much a school will
receive (see chapter 4).
Summary
There are three major impediments preventing education systems in
ECA countries from helping to reduce the skills shortage in their region.
First, ECA education systems have been operating without crucial
information—in the dark. By focusing on top-achieving students, they
neither seek to deliver quality education to all students, nor do they sys-
tematically collect data on the learning and employment outcomes of all
students. Second, the legacy of central planning keeps the governance and
management of school systems highly centralized, with central policy
makers intensely involved in operational details. A focus on compliance
with norms, together with financing schemes based on inputs and not
outputs, means that most local education authorities and school princi-
pals in the ECA region lack the autonomy and authority to make crucial
management decisions for their own institutions.
Third, education systems in the ECA region use financial resources
highly inefficiently. Nowhere is this more apparent than in the pre-
university sector, where few countries have adjusted teacher staffing levels
in response to falling student numbers over the past 20 years. As a result,
student-teacher ratios have fallen sharply as per student costs have ris-
en—more so than in any other region in the world.
Policy makers in the region can overcome these impediments by com-
mitting to quality education for all students, supported by systematic data
collection on student outcomes; expanding autonomy at lower levels of
the education system in return for accountability for student perfor-
mance; and introducing performance-based financing. ECA countries are
specifically encouraged to develop national learning assessments, system-
atically participate in international assessments, conduct regular tracer
studies of graduates to determine their employment outcomes, and then
use this data to inform education policy.
Overcoming the legacy of central planning will require central policy
makers to reduce the role of the central government in day-to-day
operations of their education systems. By concentrating on setting system
standards and overall goals, the central government can then manage
112 Skills, Not Just Diplomas
Notes
1. As a nontransition country, Turkey sometimes represents a different case.
Although much of the discussion in this chapter is relevant to Turkey, it will
be noted in those cases where it is not.
2. Although Turkey was not a command economy, it too has a very centralized
system, evident, for example, in the lack of autonomy and accountability in
the education system.
3. Again, though Turkey was not a command economy, it has a very centralized
education system.
4. The PISA assessment conducted by the OECD provides evidence of these
large differences in school autonomy, based on a questionnaire administered
to school principals (in which three questions relate to budgeting). Data from
the 2006 PISA show that schools in Estonia enjoy the most autonomy of all
ECA countries, irrespective of how the question on budgeting is asked, while
schools in Azerbaijan enjoy the least. It should be noted that PISA data must
be interpreted with some caution, as it is unclear what is meant by “having
more autonomy on budgeting.” Does it mean more autonomy to formulate
budgets, to re-allocate resources within budgets, to determine salaries and
annual increases, or something else? Still, certain ECA countries have unam-
biguously granted substantially more flexibility to schools, principals, and
governing boards than have others. A much better source of information on
differences in school autonomy is provided by the Teaching and Learning
International Survey (TALIS) study, also conducted by the OECD. In this
survey, teachers and principals are asked more specific questions related to the
degree of autonomy in their schools. Unfortunately, only six ECA countries
Resolving the Skills Shortage in the ECA Region: A Policy Framework 113
(Turkey, plus all new EU member states at the time) participated in the 2008
study, which confirmed that schools in Estonia enjoy substantial autonomy.
5. Comparing teacher salaries to those of other professions is not an easy task,
both because salary differentials can derive from differences in educational
background, experience, and other factors, and because different professions
enjoy different noncash benefits. However, available data show that teacher
salaries in the ECA region are low and that the gap between their salaries and
the median economy-wide salary has widened in many countries. See the fol-
lowing data sources on teacher salaries: for OECD countries, OECD 2008
(accessible online at www.oecd.org/edu/eag2010); for CIS countries, CISSTAT,
online database of the Interstate Commerce Committee of the Commonwealth
of Independent States (Statistics of the Countries of the CIS, Annual Data
from 1991 to 2010, http://www.cisstat.com/2base/frame00.htm); for economy-
wide annual salaries for all countries, OECD StatExtracts, http://stats.oecd
.org/Index.aspx. (All URLs accessed December 2010.)
6. The “result” referred to in this figure is a student enrolled. As discussed in
chapter 4, a more desirable per student financing scheme would be to finance
a student who is graduating, or better yet, a student who is graduating with a
desired level of competencies. However, even the most advanced ECA coun-
tries (in terms of moving to school financing based on results) still base
financing on the number of students enrolled (an intermediate result, at best).
This report therefore intentionally distinguishes between the inputs of teach-
ers, classes, and finances, and among students “enrolled,” “graduating,” “gradu-
ating with a desired level of competencies,” or “graduating and finding a job”
as results (or outputs).
CHAPTER 4
115
116 Skills, Not Just Diplomas
(if any) countries in ECA are yet using the results of these exams to help
design educational policies. To do so effectively, several key actions are
needed. Most importantly, some background information about the stu-
dents who take national assessment exams needs to be collected. The
importance of this information should not be underemphasized: in its
absence, the usefulness of test scores is severely constrained.
In Bulgaria, for example, a national assessment center can generate
average national and regional scores on student achievement tests, but
this data cannot be disaggregated to identify the performance of either
different groups of students or individual students. The problem is that
the assessment system (which gathers data on student test scores) is not
linked to either socioeconomic information about students (e.g., data
collected via survey when students take the test) or to the individual
student identification numbers assigned by the Ministry’s education
information system. This oversight means that education policy makers
in Bulgaria are unable to answer crucial questions related to the large
wave of school closures that took place during the summers of 2007–
2009, such as whether students who were moved to new schools are
doing better or worse academically.
Further recommendations to improve assessment systems include the
need to administer such tests on an annual basis, not only to collect stan-
dardized longitudinal data, but also to enable countries to have a baseline
of student achievement results before and after educational reforms are
introduced. Moreover, countries need to move beyond using national
assessments solely for examination purposes (that is, to certify that a stu-
dent has mastered a curriculum at the end of compulsory and upper
secondary education) and begin using such tests to assess the quality of
education at different stages in the cycle. For instance, policy makers may
want to use a standardized test administered each year in order to evalu-
ate how well fourth-graders are reading.
Box 4.1
In Romania, researchers used PIRLS 2001 and 2006 data to quantify the extent
to which students from rural schools lagged behind their peers from urban schools
(Romania 2009). Equally important, they pointed out that the gap had widened, not
narrowed, between the two tests. The magnitude of the gap played an important
role in triggering the design of a project (financed by the World Bank) to address the
quality of education in rural areas (see World Bank 2003b).
(continued next page)
Managing for Results at the Pre-University Level of Education 119
Finally, to gauge the risk that tertiary students were withdrawing from school
as a result of the ongoing financial crisis, researchers in Romania administered a
survey of students to identify those most at risk of dropping out (Pricopie et al.
2010). These results can be used by policy makers to design policies to mitigate
this risk.
examine their educational progress over time because the tests are com-
patible and administered at pre-determined time intervals (e.g., PISA is
administered every three years; PIRLS, every five years).
Create demand for more data and analysis. In order to mobilize and
sustain resources to collect and analyze data, demand for this data and
analysis must be stimulated. If such data are not used in the policy process,
moreover, policy makers are quick to axe resources allocated for these
purposes. Given the weak internal capacity of most ministries of educa-
tion in the region, central governments might consider relying on outside
researchers—such as university and policy institute experts—for help. This
could be done in a number of different ways, including the following:
120 Skills, Not Just Diplomas
• Set aside competitive grants for researchers to write policy notes and
research papers using assessment data.
• Make entire datasets more readily available. For instance, the OECD
makes the entire PISA dataset publicly available for researchers to
download online (http://pisa2009.acer.edu.au/; accessed December
2010).
• Change policies to make data matter, such as making improved student
learning outcomes a component of teacher and school evaluations. (If
the results of assessments have no impact, whether directly or indi-
rectly, on policies or people’s careers, it is very difficult to mobilize and
sustain resources to maintain assessment centers and train people to
analyze results.)
• Create and empower school councils and create incentives for their
active participation in school decision-making processes, and provide
them with adequate capacity and training, to increase the participation
of parents and the community (see Patrinos et al. 2010).
• Administer national student assessments at key stages of the education
cycle (e.g., after the 4th, 7th, and 12th grades) and make this data read-
Managing for Results at the Pre-University Level of Education 121
ily available to parents and school councils to help hold schools ac-
countable for results. Making assessment results more readily available
has helped improve state-level accountability and student achievement
in the United States (Hanushek and Raymond 2005; Carnoy and Loeb
2002) and Mexico (Alvarez et al. 2007).
• Require schools to prepare school development plans that outline each
school’s strengths and weaknesses, together with a vision and action
plan for making improvements. Ideally this exercise should take as its
starting point student scores on a national assessment exam, together
with an institutional self-evaluation.
• Prepare “school scorecards” for all schools that include basic indicators
(whichever are available), giving parents and students a sense of a
school’s performance. Such scorecards should be made available on the
ministry’s website and disseminated to local authorities and schools.
Such score cards should publish the “value-added” of a school (e.g., an
increase in test scores that takes into consideration students’ socioeco-
nomic background).1
• (Ideally) disaggregate school score cards—where regulations permit—
by the performance of
° Minority students
° Low-income students
° Students with limited proficiency in the language of instruction
° Students with special education needs or disabilities or both
• Agree on criteria that objectively identify “a school in need of improve-
ment” and decide what actions will be taken to support such schools.
More specifically, they can agree on the ultimate sanctions for contin-
ued failure to improve performance, such as
° Re-constituting the school (i.e., formally closing it down but reopen-
ing it immediately with new management and the possibility of staff
re-applying for their jobs).
° Turning the school over to private management
• Provide rewards to high-performing or improving schools or both. Here
policy makers should avoid characterizing a “high-performing school”
simply as one that has high average student scores on a standardized
test or a large number of graduates who go on to university, as these
indicators may merely reflect the fact that the school attracts high-
caliber students (possibly from well-off families). Rather, a school
122 Skills, Not Just Diplomas
placed on probation for two to four years, during which time they receive
support and guidance to improve student achievement.2
Box 4.2
Box 4.3
autonomy at the tertiary level in the ECA region has not been without
problems. Unfortunately, unethical behavior is flourishing at the tertiary
level, which undermines the sector’s ability to deliver quality education
for all.
There are, however, ways of moving forward at the pre-university level
while addressing these concerns. For instance, institutional autonomy
does not have to be granted to all schools at the same time. Since central
authorities might reasonably have concerns about granting autonomy for
management when capacity at lower levels is very limited, one practical
approach would be to start with larger schools (and larger local educa-
tional authorities), which usually have more managerial experience. From
an implementation perspective, the benefits of focusing on large schools
are obvious. In the case of Romania, for example, focusing initially only
on large schools would involve training, say, 1,000 school principals and
accountants, rather than the principals and accountants of more than
7,000 small schools.
The degree of autonomy granted to schools can also be expanded
gradually, with country circumstances influencing the ultimate degree of
autonomy eventually granted. It is important to note that autonomy—
like accountability—is complicated. For instance, granting schools more
autonomy in budgetary matters is unlikely to have much impact unless
this autonomy includes the ability to make staffing decisions (since staff
costs are the single largest budget item). Similarly, autonomy over staffing
decisions will, inevitably, involve providing schools the autonomy to
determine class sizes. Policy makers are thus urged to simultaneously
make changes to a number of rules and regulations as part of a change in
class sizes.
Box 4.4
receive less funding (in nominal terms) under the per capita scheme than they
had received the year before the new scheme was introduced (i.e., in 2006). To
allow these municipalities time to adjust their class sizes (and, if needed, reduce
the number of schools), they were entitled to receive compensation equivalent
to the difference between the allocation based on the new per student amount
and their allocation in 2006 during the initial year of implementation.a
However, to ensure that these “losing” municipalities would eventually be able
to operate in a fiscally sustainably manner under the new regime, they were
required to present a school consolidation plan to MES and the Ministry of Finance
in order to qualify for the compensation. In addition, the MES clearly communi-
cated that the compensation would only be available during the first year of im-
plementation. Thus, losing municipalities had strong incentives to initiate school
consolidation; with wages increasing by about 5–10 percent per year (at the time)
in nominal terms, the only way for them to become financially viable under the
per capita scheme was to consolidate their school networks. As expected, the
reforms led to a large number of school closures, as municipalities moved to cre-
ate more viable schools.
Source: Authors based on World Bank 2007b.
a. On the other hand, implementation plans did not specify any ceilings on funding for municipalities
that would be “winners” under the new scheme. Thus, municipalities that had well-optimized school
systems would experience increases in their education budgets.
35
30
25
20
15
10
0
Slovenia
Estonia
Slovak Republic
Poland
Hungary
Czech Republic
Bulgaria
Azerbaijan
Armenia
Ukraine
Russian Federation
Turkey
Serbia
Kosovo
Luxembourg
Greece
Spain
Finland
Germany
France
Ireland
United Kingdom
Mexico
United States
Brazil
Chile
Korea
OECD Average
EU10+1 LI CIS MI CIS Turkey Balkans Western Europe other
Sources: OECD Education At a Glance Database for OECD and partner countries; authors’ calculations based on Ministry of Education data (from education management information
systems) for Azerbaijan, Serbia, and Romania, and on National Statistical Institute school-level data for Bulgaria.
Note: Most recent data for each country (2005, 2006, 2007, or 2008).
131
132 Skills, Not Just Diplomas
Figure 4.2 Average Class Size in Large Primary and Secondary Schools in Nine
ECA Countries, Various Years
35
average class size (number of students)
30
25
20
15
10
0
9)
8)
7)
9)
9)
9)
9)
7)
10
00
00
00
00
00
00
00
00
20
(2
(2
(2
(2
(2
(2
(2
(2
a(
va
ia
ia
ia
ia
ia
ni
in
ija
rb
ar
tv
an
en
do
to
ra
ba
lg
La
Se
m
ol
Uk
Es
Bu
er
Ro
Ar
M
Az
Source: Authors’ analysis based on school-level data obtained from the ministries of education or national statis-
tical agencies of nine ECA countries for various years.
Note: The definition of a “large” school is flexible in order to recognize different types of schools in ECA countries.
A primary school that teaches only grades 1–4 is defined as a large school if it has more than 50 students per
grade and thus a total of more than 200 students. A comprehensive general school that teaches grades 1–12 is
defined as a large school if it has more than 600 students.
Address Overstaffing
School systems in the ECA region have to date focused more on protect-
ing teaching positions than on ensuring fewer, higher paid jobs. In light of
mounting cost pressures, many countries in the region are now trying to
downsize the number of both schools and teachers in their systems. In
addition to increasing class sizes and optimizing existing schools, coun-
tries in the region might explore the scope for retiring over-age teachers.
Specifically, they could use the impending wave of teacher retirements to
implement an orderly reduction in the number of teaching positions,
Managing for Results at the Pre-University Level of Education 133
Box 4.5
300 3500
number of school closures
3000
250
number of schools
2500
200
2000
150
1500
100
1000
50 500
0 0
2
19 –94
19 –95
19 –96
19 –97
19 –98
19 –99
20 –00
20 01
20 02
20 –03
20 –04
20 05
20 6
20 –07
20 08
9
3
–9
–0
–0
–9
–
–
–
91
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
92
19
19
19
13.2
student-teacher ratio
22.0
13.0
21.8 12.8
21.6 12.6
12.4
21.4
12.2
21.2
12.0
21.0 11.8
00
01
02
03
04
05
06
07
80
20
20
20
20
20
20
20
20
20
Source: Authors’ calculations based on school-level data obtained from Bulgaria’s National Statistics Insti-
tute Web site, www.nsi.bg/index_en.htm.
Managing for Results at the Pre-University Level of Education 135
while slotting new hires for subjects in greatest demand, such as informa-
tion technology. The use of stronger redundancy or retraining packages or
both for departing teachers might make departures more acceptable. The
experience of countries now in the process of reducing teacher numbers
in response to the ongoing economic crisis (e.g., Latvia and Romania)
may provide valuable lessons in this area.
At the same time, governments in the region will need to develop
strategic policies to reinvigorate the teaching profession (see box 4.6).
Box 4.6
Summary
To manage for results at the pre-university level, education ministries
should implement national student assessment exams on an annual
basis, as well as participate regularly in student assessments such as
PISA, TIMSS, and PIRLS. For assessment data to produce the kind of
information needed to improve student outcomes, however, background
information on the students who take these assessments must be col-
lected at the time of the tests. Without such information, the usefulness
of test scores is severely constrained because the data cannot be disag-
gregated by different types of students, such as those from different
Managing for Results at the Pre-University Level of Education 137
Notes
1. For an example of school scorecards, see the Ministry of Education of Chile
website at www.simce.cl. For an example of school report cards with value-
added results, see U.K. school results on the U.K. Department for Education
website at www.education.gov.uk/performancetables./ (both URLs accessed
January 2011).
2. U.S. Department of Education. 1998. “Turning Around Low-Performing
Schools: A Guide for State and Local Leaders.” U.S. Department of Education,
Washington, DC. http://www.ed.gov/pubs/turning/intervene.html. Accessed
September 2010.
CHAPTER 5
139
140 Skills, Not Just Diplomas
Governments are not the only actors that would benefit from better
information on outcomes. Employers, students, and parents also need
objective, standardized data to make informed decisions. The market for
tertiary education is a classic example of the seller (i.e., universities) being
vastly better informed than their consumers (i.e., students, parents, and
future employers), or even their financiers and donors (i.e., governments,
for the most part) (Carey 2010). When students make choices about
future universities and degrees, they usually do so without having met
the professors who will teach them and with little to no understanding
of the skills that they will acquire. This asymmetry places a great deal
of power in the hands of the seller, which can contribute to worsening
quality and higher costs for the consumer. In other parts of an economy
characterized by similar informational asymmetries, policy makers usu-
ally play a strong regulatory role or mandate that providers provide
better information to consumers or both.
Currently, data on tertiary outcomes in the ECA region as elsewhere
focus on research outcomes, such as how many publications the faculty of
a university is producing. Unfortunately, when this is the only performance
indicator available, it inadvertently receives more attention than it should.
Indeed, rankings of universities in the region place significant weight on
the number of publications that they produce. However, from a skills per-
spective, the outcome that matters most is whether students are graduat-
ing with the competencies in demand on the job market and finding jobs.
Currently, these outcomes are not being measured. Governments in the
ECA region can change this situation by granting tertiary institutions pub-
lic resources in return for them providing useful information about stu-
dent learning and employment outcomes. There are several complementary
ways of gathering this information, some of which are more direct than
others. These include rankings or league tables, tracer studies, and direct
measurement of student competencies via standardized tests.
As noted in earlier chapters, standardized tests are the best available
indicator for measuring student performance, especially when socioeco-
nomic characteristics are collected for each student (allowing the results
of different socioeconomic groups to be compared). Such data are espe-
cially useful if organized to track the progress of individual students over
time. Together with other indicators, these tests provide data on differ-
ences in learning outcomes between weaker and stronger students and
institutions. Without this information, policy makers have no way of
identifying—and thereby rewarding—institutions that do the best job of
giving the largest number of students a quality education. Unfortunately,
142 Skills, Not Just Diplomas
Box 5.1
Box 5.2
Box 5.3
not only by this law and the statute of the institution, but also by other laws. In
practice, this meant that universities could not own their buildings or borrow
funds, and were only able to spend budgets as they saw fit to a limited extent. As
part of the 2009 reforms, these inconsistencies are being addressed by changing
the legal status of universities from a budgetary to a public entity. This will provide
them more freedom and the right to own property, as well as expand their rights
to manage the property entrusted to them by the state.
Strengthen Accountability
Accountability at the most basic level refers to fundamental academic
and fiscal integrity, which touches on such issues as admissions and finan-
cial fraud, plagiarism, and professorial nepotism. Here, almost all ECA
countries continue to have serious problems. Strengthening accountabil-
ity in tertiary education also entails ensuring that public resources are
spent effectively, the education provided is of high quality, and that study
programs are relevant to students’ future in the workforce. Although it
might seem necessary to address basic integrity issues before dealing with
higher-order accountability issues, all of these problems can be addressed
concurrently, a task made easier by the fact that most basic accountability
instruments also address higher-level concerns.
Table 5.1 Tools for Strengthening Basic Academic and Fiscal Integrity in
University-Level Institutions
Tools that strengthen academic integrity Tools that strengthen fiscal integrity
• Licensing • Financial audits
. . . and educational quality . . .and fiscal efficiency
• Accreditation, academic audits, and • Public disclosure laws
evaluations
• Public disclosure laws • Strategic budget plans
• Fostering of outsiders to review academic • Performance contracts based on
integrity (e.g., Romania) performance indicators
• Performance contracts based on • Student loans, scholarships, and vouchers
performance indicators
• Creation and dissemination of rankings, • Embedding of incentives into allocated
benchmarks resources (e.g., tying funding to the
number of students enrolled or graduated,
rewarding good performers, etc.)
Sources: Authors’ review of available instruments, as presented in Salmi 2009.
tool been used in Romania, see box 5.4). Performance contracts based
on performance indicators can also help ensure integrity and educational
quality. Finally, it can be valuable to facilitate the creation of rankings or
benchmarks or both and then disseminate these data in an appropriate
manner.
Instruments that address basic fiscal integrity also have an impact on
the efficient use of resources. One obvious tool for improving basic finan-
cial integrity is the financial audit. Tools that accomplish both fiscal integ-
rity and the efficient use of resources include public disclosure laws,
strategic budget plans, performance contracts, external watchdogs (see
box 5.4), and student loans, scholarships, and vouchers. In addition, it can
be useful to embed incentives into the resources allocated to higher edu-
cation institutions, for example, by linking funding to the number of
students enrolled or graduating or using it to reward good performers.
These suggestions are discussed in greater detail in the section on financ-
ing below.
Box 5.4
• Tertiary institutions have primary responsibility for quality and are re-
quired to have processes of internal quality assurance.
• Tertiary institutions are subject to external oversight by an agency
charged by the government with assuring the competency of quality
assurance mechanisms.
• Quality review agencies themselves submit (on a voluntary basis) to
quality assessment procedures through the European Quality Assur-
ance Register, which is a joint project of the European Network of
Quality Assurance Agencies (ENQA), together with the European
Managing for Results in the Tertiary Education Sector 157
Box 5.5
fully functional
system of QA with
peer review of QA
functional system
agency, and
of QA covering all
significant student
institutions and with
and international
beginning of a substantial student
participation in
national QA and international
QA process
system with some participation
student and in QA
international
not started or in participation in
early stages of Croatia
QA process
development of Czech Republic
quality assurance Albania Hungary
Armenia Latvia
Bulgaria Lithuania
Estonia Poland
Georgia Romania
Azerbaijan Moldova Macedonia, FYR
Bosnia and Russian Federation Turkey
Herzegovina Serbia
Montenegro Slovenia
Slovak Republic Ukraine
Performance Contracts
The introduction of performance-based budgeting can be a powerful
instrument that ministries of education (and finance) can use to guide
the sector, as it causes governments to shift from line-item to out-
come-based funding (OECD 2007c). At the same time, universities
are given greater autonomy in how they spend their budgets and are
held accountable for delivering results. Three models of performance-
based funding are currently being used in higher education (Ziegele
2009): (1) formula funding, (2) target agreements, and (3) competi-
tive funding (see table 5.2).
A recent report (OECD 2007c) reviewed the experiences of OECD
countries with performance-based budgeting and highlighted five poten-
tial benefits. First, this type of budgeting generates a sharper focus on
results within the public sector. Second, it provides more and better infor-
mation on government goals and priorities, as well as how different pro-
grams contribute to these goals. Third, it encourages greater emphasis on
planning and signals what is working and what is not. Fourth, it improves
transparency by providing greater and more useful information to the
central government and the public. Finally, it has the potential to improve
both the efficiency and the management of individual programs.
Managing for Results in the Tertiary Education Sector 161
Summary
In terms of the relationship between the central government and tertiary
institutions, ECA countries today can, broadly speaking, be divided into
two groups. The first group consists of most new EU member countries,
which have already expanded the autonomy of their tertiary institutions.
Their main challenge today is to strengthen a still weak accountability
structure, evident in persistent allegations of corruption in the sector and
doubts as to whether existing accountability mechanisms (such as sanc-
tions, accreditation agencies, and so forth) are sufficiently strong to induce
providers to focus on improving learning and employment outcomes
(as opposed to other goals of tertiary institutions). The second group of
countries—mostly low-income CIS countries—have maintained tight
Managing for Results in the Tertiary Education Sector 163
Notes
1. Given the complexity of the tertiary sector, some clarifications are needed.
First, the “providers” discussed in this chapter, both public and private, are all
institutions that offer tertiary education programs. Second, it is readily
acknowledged that tertiary education institutions serve important societal
goals that stretch far beyond that of providing graduates with employable
competencies. For instance, through their research, these institutions help
expand a society’s collective knowledge and can contribute to innovations in
the economy, giving students employable competencies in the process. And
through their political independence, universities help society hold public
164 Skills, Not Just Diplomas
officials accountable. Yet, the focus of this book—the need to create educa-
tion systems that are more responsive to labor market needs—implies that
these broader goals of tertiary institutions (and the policies related to enhanc-
ing their performance in those dimensions) are beyond its scope.
2. See a recent report on university autonomy by the European University
Association (Estermann and Nokkala 2009). The EUA report does a care-
ful job of discussing the complexities and all dimensions of university
autonomy.
3. The Bologna Process was kicked off by the signing of the Bologna Declaration
in 1999.
4. See Bologna Working Group 2005 for further details.
5. For additional information on the Tuning Project, or Tuning Educational
Structures in Europe, visit its official website at http://tuning.unideusto.org/
tuningeu/ (accessed September 2010).
6. See the online guide at www.guardian.co.uk/education/universityguide/
(accessed January 2011).
7. For a discussion of rankings, Kehm and Stensaker 2009.
8. The scorecard is the result of a regular stock-taking exercise undergone by
countries that participate in the Bologna Process. The last such stock taking
was carried out in 2009.
9. The “European Standards and Guidelines for Quality Assurance in the
European Higher Education Area” is available on the website of the European
Quality Assurance Register for Higher Education at http://www.eqar.eu
(accessed January 2011).
10. In addition to scoring countries on three indicators related to quality assur-
ance (“external quality assurance,” “student participation in quality assurance,”
and “international participation in quality assurance,” the 2009 stock-taking
exercise included a “qualitative analysis” of various aspects of internal quality
assurance.
11. Private expenditures on higher education are somewhat difficult to capture.
Data from the UNESCO Institute for Statistics on this subject is quite
inconsistent, in part because it includes income from many different sources
(see the UIS website, http://www.uis.unesco.org/; accessed September 2010).
However, it is possible to look at fee income in a relatively straightforward
way, based on accessible published sources. In most of the OECD region
(with the exception of Japan, the United States, and Mexico), for example,
fee income accrues almost entirely to public educational institutions.
CHAPTER 6
Lifelong learning and adult education and training are fast becoming an
important element of education systems across the ECA region. Although
demand for labor in the ECA countries has already shifted toward
higher-order, multifaceted competencies, it has also become more unpre-
dictable, indicating the need for a market-responsive sector that can eas-
ily facilitate the reskilling of the adult labor force. The reskilling of the
work force is needed not only to compete in the global economy, but also
to address the lag effects of the transition to market economies.
Significant segments of the labor force—often comprised of less-skilled
and more disadvantaged workers—remain inactive in many areas. Given
the rapid demographic decline expected in many ECA countries, main-
taining high rates of growth will require an increase in labor force par-
ticipation and productivity.
High-quality, relevant adult learning relies on (1) sound regulation and
quality standards that promote autonomy and accountability (largely miss-
ing in all but a few ECA countries), including recognized qualifications
frameworks and certificate programs, and (2) financing incentives that
promote a competitive market and overcome the obstacles and market
failures associated with the provision of adult learning. Overcoming infor-
mation barriers for individuals and firms is central to both policy levers.
165
166 Skills, Not Just Diplomas
creates the basis for a practical legislative and regulatory framework. The
EU, for example, has advised its member states to develop and implement
strategies on lifelong learning to advance the education and training
dimension of the Lisbon Jobs and Growth Strategy and its successor, the
Europe 2020 Strategy. The EU10 have been adopting such strategies
since 2005 (see box 6.1 for a case study).
Box 6.1
Box 6.2
(1) demand-led funding, that is, routing public funds for adult vocational skills
training through demand-led channels, thus ending supply-side planning of
skills provision;
(2) strengthening the voice of employers by rationalizing the number of bodies
that articulate the views of employers and creating a single Commission for
Employment and Skills; and
(3) providing economically valuable skills by relying on reformed and relicensed
Sector Skills Councils to ensure that vocational qualifications reflect the skills
valued by employers (Leitch Review of Skills 2006).
the system (Dearden, McGranahan, and Sianesi 2004). Yet despite the growth in
these qualifications, wage returns to NVQ Level 2 qualifications have to date
largely been low or nonexistent, with a few exceptions (Dearden, McGranahan,
and Sianesi 2004).
In addition, various projects provide information to workers to keep them bet-
ter informed about market demands and the returns to training. A number of pro-
viders also offer advice to workers, particularly those who are unemployed or
qualify as low skilled. These efforts include Jobcentre Plus work to identify the skill
needs of Jobseeker Allowance recipients who have been unemployed and inac-
tive for six months; Learndirect’s personalized advice, information, and guidance
tailored to the needs of adult clients at the skill level NVQ 2 or below; and the com-
munity outreach of Learning Ambassadors, local volunteers who have themselves
undergone adult training, to identify and work with people in similar situations.
Individual Learning Accounts (ILA) in the United Kingdom provide funding to
individuals with the lowest skill levels, making training accessible to a group that
otherwise would not participate. ILAs make a maximum sum of between £100
and £200 available to low-skilled individuals, who may spend it on a qualification
and provider of their choice. Individuals and their employers are also encouraged
to contribute to ILAs. Opening an ILA comes with a guidance package that targets
individuals who are least able to borrow from private financial institutions. In
2000, England implemented one of its largest ILA projects, with 1.4 million par-
ticipants, but it was later closed due to irregularities. The project was found to
suffer from significant deadweight loss, with more than half of participating indi-
viduals stating that they would have sought training without their ILAs (OECD
2005c). Subsequently, however, both Scotland and Wales implemented compa-
rable schemes and since 2007, England again began to pilot a similar project.
Lastly, the government’s Train to Gain Program helps employers identify train-
ing needs and then tailor corresponding training packages which are jointly or
fully financed by the government. The program offers financial incentives to
workers with paid time off and subsidies to employers to compensate for trainee
wages. This intervention targets all market failures, from financing training to over-
coming the time preferences of employees, who worry about the opportunity
cost of training. The Train to Gain initiative was first piloted in 2002–2006 and
covered 30,000 employers and 250,000 employees. It has now been mainstreamed
across the country. Initial evaluations indicate a small positive impact on the take-
up of training for both employees and employers (Abromovsky et al. 2005).
Sources: Abromovsky et al. 2005; Dearden, McGranahan, and Sianesi 2004; OECD 2005c; Leitch Review
of Skills 2006.
Advancing Adult Learning in ECA 171
Accountability
Information is the key to accountability in adult education and training.
Functioning accountability systems require government regulatory
intervention to ensure that (1) adequate information is available for
consumers of such education (so that they may make informed deci-
sions), and (2) the suppliers of adult education and training programs
maintain the quality and relevance of their program offerings. Thus,
policies that set standards for and ensure the quality of adult learning
opportunities both foster supply and stimulate demand. For example,
national qualification frameworks can help to stimulate demand by
allowing workers to obtain certified qualifications that prove their com-
petencies to employers—and address a key information constraint. They
can also make continuing vocational education and training (CVET) and
retraining of unemployed workers more strategic and relevant to eco-
nomic needs. Clearly, the regulation and certification of education pro-
viders—both public and private—and their programs is a key element of
developing an adult learning system.
National qualification frameworks (NQFs) are being developed in
many ECA countries to establish coherent quality standards for educa-
tion and training systems and recognize prior learning, including that
acquired through formal, nonformal, and informal learning methods. For
many countries in the ECA region, NQFs—at least for the tertiary level—
are a mandatory component of the Bologna process (see chapter 5). The
development of such frameworks is, however, complex. NQFs require a
strong legislative basis, a high degree of coordination and administrative
capacity, and a great deal of time—countries such as Ireland and the
United Kingdom have spent years developing and refining them (Gunny
and Viertel 2007; Ireland 2003; United Kingdom 2003). Moreover, they
are no panacea, and it is important to set realistic expectations as to what
they can achieve in conjunction with other policy levers in adult educa-
tion and training. A recent assessment of the experience with NQFs in
16 countries worldwide showed that they may not always result in the
better matching of demand for and supply of adult education and train-
ing, nor in better training outcomes (Allais 2010). The success and failure
of NQFs appear to depend, at least in part, on the extent of the construc-
tive involvement of key stakeholders—and their willingness to eventually
make use of frameworks devised.
As noted in the previous section, employer-led, sector-level councils
often play a critical role in developing standards. For example, in the
Czech Republic, sector skills councils have started to develop national
174 Skills, Not Just Diplomas
developed institutes that evaluate adult learning programs with the aim
of ensuring quality. The United Kingdom has established an Adult
Learning Inspectorate that assesses hundreds of publicly funded programs
every year and publishes the results; those that underperform often lose
funding (OECD 2005c). Quality assurance systems in adult education
and training can also involve the use of self-evaluations by providers
alongside external evaluations, as in the case of Slovenia. In the Czech
Republic, the Ministry of Education, Youth, and Sport certifies retraining
programs for the unemployed that are financed by the Labor Office. The
ministry is also beginning to pilot a quality assessment system and plans
to institute a star rating scheme for quality based on a simple set of crite-
ria, rather than complicated quality standards. If successful, this could be
an interesting option for less advanced countries.
In ECA countries, programs that assess the quality of adult education
and training may play an important role in helping adults, particularly
those with little experience of these programs, make decisions. As noted
in box 6.3, cost is a major barrier to adult education and training in the
region: Polish providers have determined that learners are likely to
choose a program based only on price, irrespective of quality (OECD
2005d). For providers that do not typically receive public funding, qual-
ity seals are an attractive means of assessment. Even though such seals
rely on voluntary participation, providers have the incentive to meet
standards in order to demonstrate the quality of their product to the
market. Both Germany and Austria have developed quality seals. In
Austria, the initiative was originally supported by nonprofit providers
and in Germany by market providers, but both receive funds from the
government (OECD 2005c).
Box 6.3
60
that this was a constraint
50
40
30
20
10
0
A
N
T
P
U
P
L
R
R
T
C
U
E
R
ES
PO
ES
AU
CY
SW
SV
IT
UK
BG
NO
FI
GR
LV
LT
DE
HU
other EU EU 10
SWE
70
60
FIN
50 UKR
DEU SVK
40 CYP AUT
BGR
LTU FRA
LVA
ESP
30
20 ITA POL
GRC
10 HUN
0
0 10 20 30 40 50 60
% of respondents reporting that information regarding
training opportunities is available
training because they cannot be sure that they can demonstrate their raised
productivity to employers. Conversely, employers and workers in developing
economies may not be fully aware of the productivity returns from training due
to lack of information about the type and area of training, as well as a lack of
managerial skills in small and medium-size enterprises. Survey data from ECA
confirm a positive correlation between the incidence of training and the avail-
ability of information (see figure B6.3, panel b), with the information barrier par-
ticularly acute for less educated workers.
Source: Authors.
178 Skills, Not Just Diplomas
Box 6.4
When focused on basic skill gaps, these programs can also promote social
inclusion (see box 6.5).
Although many ECA governments have long funded industrial
training institutes that trained or retrained unemployed adult workers,
in most instances, these institutes were not always designed to meet the
needs of the modern work force. Governments today would be better
Box 6.5
90
80
70
60
percent
50
40
30
20
10
0
ar m nd
ar m ing
ar m ing
gu ploy I fu
te t
te t
te t
gu ploy ain
gu ploy ain
an en
an en
an en
e)
e)
e)
U
r
m t
m t
em ing
(e lass
(e -job
(n rain
-c
e
-th
in
t
ss
o
on
- cla
in
Source: Staff calculations based on administrative data for 2008 provided by the Turkish Public Employment
Service (ISKUR).
Advancing Adult Learning in ECA 187
Figure 6.2 Adult Education and Training Priorities in the ECA Region
group 1: less
advanced, with group 2:
20
demographic crisis: advanced, with
consider demographic
UKR BGR crisis; priority
BLR RUS
GEO
10 ROM LVA
LTU
percentage of population decline 2010–30
MDA HRV
H V
BIH POL N EST
HUN
ARM CZE SVN
SV
MKD SVK
0 SLB
MNG
ALB KAZ
–10 AZE
KGZ
–20 TUR
UZB
group 3: less
–30 advanced, with no
demographic crisis:
TJK
no priority
–40
0 5,000 10,000 15,000 20,000 25,000 30,000
GDP per capita (2007 US$)
Turkey and Kazakhstan do not fit well into any of these categories, but
it can be argued that their priorities are the same as those of group 1.
Table 6.1 provides a rough policy framework for adult education and
training in ECA countries, based on their respective economic and demo-
graphic indicators.
Table 6.1 Recommended Policy Framework for Adult Education and Training in
the ECA Region
Advanced Less advanced Less advanced
economies with economies with economies with
demographic demographic no demographic
challenge challenge challenge
ARM, BIH, BLR,
GEO, MDA, MKD, ALB, AZE, KGZ,
EU10, HRV, RUS MNE, SRB, UKR TJK, UZB
Policy and institutional foundation
Multistakeholder skills task force
National strategy on adult
learning
Coordination mechanisms
Autonomy
Promote autonomy of training
institutions
Accountability
National qualifications framework
Regulation and certification
regime
Certification and quality control
agency
Strategic financing
CVET—firms
Training tax deductions for firms
Training grant schemes based on
payroll levy
Grant schemes
SME support schemes, with training
focus
CVET—individuals
Worker tax deductions for
training
Subsidies or vouchers for
individuals
Training leave
Individual learning accounts
Loan schemes for individuals
Retraining and education for the
nonemployed
Job training for the unemployed
Second-chance education
programs
Source: Author’s assessment.
Note: Color key: priority, consider, not a priority. See “Acronyms” for country abbreviations and groupings.
190 Skills, Not Just Diplomas
Summary
Lifelong learning and adult education and training are growing in
importance across the ECA region. Increasing demand for skilled labor,
accelerating technological change, and significant demographic decline
are all contributing factors. Although building adult learning systems is
critically important, its overall priority in individual ECA countries
depends on economic and demographic conditions and their most
pressing education challenges.
Successful adult education and training systems are based on a high
degree of policy coordination and partnership between government
agencies and the private sector, with a strong voice for the demand
side—businesses and individuals—in determining training content. Once
a solid policy foundation has been created, governments can “steer” adult
education and training through tracking data on program quality
(derived from both routine monitoring and evaluation and impact stud-
ies), promoting autonomy and accountability, and ensuring efficient
government financing.
Across the board, private investment lies at the heart of adult educa-
tion and training, though governments will need to provide a certain
amount of financing to overcome market failures. In the case of CVET
programs for the employed, governments will mainly need to create an
enabling environment for private investment through appropriate regu-
lation and the provision of information. In addition, various financial
incentives for more private investment can be targeted to both firms and
workers. For the unemployed, retraining and education programs are an
important part of active labor market policies in most ECA countries; in
this case, governments can use their role as purchasers of retraining ser-
vices to shape the overall adult training market.
Notes
1. For a discussion of regional variation in SEE countries, see Gunny and Viertel
(2007). See also OECD (2005d) and Hungary, Ministry of Education and
Culture (2006).
2. The program also supports SMEs with training needs analysis and provides
opportunities for group training with other SMEs.
192 Skills, Not Just Diplomas
3. Despite its evident effectiveness, Turkey’s new Law on Labor Force Training,
adopted in 2008, abolished the placement incentive scheme.
4. Bulgaria, Romania, and Belarus lie close together in terms of GDP per capita
and the scale of their demographic decline. However, Bulgaria and Romania
should consider adult education and training more of a priority than should
Belarus, given the former countries’ membership in the EU, participation in
the Lisbon Agenda, and access to financing for lifelong learning from the
ESF.
CHAPTER 7
The countries of Europe and Central Asia are coming out of a deep eco-
nomic crisis—the worst since the economic transition-induced recession
of the early 1990s and the deepest recession among all developing
regions. Post-crisis conditions look very different from those of preceding
years. Previously, growth in the region was fueled by cheap borrowing
from abroad, growing exports of goods and services (including labor ser-
vices), and, in the case of natural resource exporters, rising global demand.
The world of 2011 is a changed one. Borrowing is significantly more
expensive and export growth is restrained by potentially slower growth
in destination countries. Restoring and sustaining growth in this context
requires boosting competitiveness and increasing labor productivity—the
very reforms that were postponed in the boom years. These reforms are
all the more important given a shrinking working-age population in many
countries of the region.
The boom years also exposed significant bottlenecks to growth in the
ECA region, particularly with respect to the skills of the labor force
(Mitra, Selowsky, and Zalduendo 2010). Paradoxically for a region with
relatively high educational attainment, a shortage of worker skills
emerged as one of the two most important constraints to firm expansion
(the other being the tax regime). This finding makes it crucial for
193
194 Skills, Not Just Diplomas
are above 90 percent and primary completion rates are also very high
throughout the region. Although secondary enrollment rates in many
countries suffered a decline in the 1990s, they have since climbed and are
now at or above what might be expected in all ECA countries, given their
respective income levels.
The socialist legacy is particularly visible in low-income countries in
the region, which have the highest secondary enrollments in the world for
their income level. Tertiary enrollments, which have grown rapidly in the
past two decades, are also high relative to income levels, with the few
exceptions of certain low-income CIS countries, such as Azerbaijan and
Uzbekistan.
In addition to attainment, ECA countries also do well in delivering a
relatively good-quality education, particularly in the lower grades. In fact,
many younger students in the region outperform their Western European
students in reading at the fourth-grade level, scoring above the scale aver-
age on the Progress in International Reading Literacy Study (PIRLS) in
both 2001 and 2006.1 These results are not uniform across ECA, how-
ever. Countries that have not joined the European Union (EU), for
example, scored below the average, with the marked exception of the
Russian Federation, which scored well above the average and demon-
strated the greatest improvement between 2001 and 2006 (see figure
7.1). The commendable performance in reading at the elementary level
indicates that schools in ECA countries are very capable of providing a
quality education at the early stages.
Many ECA countries, though certainly not all, continue to do well on
international reading assessments such as PISA and TIMSS at or near the
end of lower secondary education, but only relative to the performance
of other countries at the same income level. Figure 7.2 provides a scatter
plot of 2008 per capita GDP against the PISA 2009 reading scores of
countries participating in PISA (with ECA countries highlighted by their
letter acronyms). The figure shows a positive relationship between the
two values, suggesting that 15-year-olds in countries with higher national
incomes tend to perform better on the reading component of the PISA.
Significantly, the ECA region seems divided. Many of the new EU mem-
ber states (including Poland, but also Turkey and Serbia) lie above the
regression line, demonstrating that their students apparently read better
than would be predicted on the basis of GDP per capita alone. Others lie
below the regression line, including Romania, Montenegro, Bulgaria,
Azerbaijan, and the Kyrgyz Republic. Students in these countries scored
lower on the reading component of the PISA than their GDP per capita
would have predicted.
196
Figure 7.1 Assessing a Skill—Reading Performance in the Fourth Grade: PIRLS Performance of ECA Countries, 2001 and 2006
600
550
mean performance
500
450
400
350
300
Macedonia, FYR
Bulgaria
Czech Republic
Hungary
Latvia
Lithuania
Poland
Romania
Slovak Republic
Slovenia
Georgia
Moldova
Russian Federation
Turkey
Balkans EU11 LI CIS MI CIS Turkey
Sources: World Bank staff calculations based on PIRLS test score data downloaded from EdStats Database.
Note: LI = low-income, MI = middle income.
Extended Summary: The Path for Education Reforms in the ECA Region 197
550
POL EST
2009 mean performance
500 HUN
LVA SVN
TUR HRV CZE
LTU SVK
450 RUS
SRB
BGR
ROM
MNE
400 KAZ
ALB
AZE
350
KGZ
300
0 20,000 40,000 60,000 80,000 100,000
2008 GDP per capita, PPP (current international $)
Sources: PISA 2009 Database and World Bank staff calculations based on World Development Indicators Database.
Note: The figure shows a regression line representing countries’ predicted PISA reading scores based solely on
GDP per capita, compared to the OECD mean reading score (horizontal line) on the assessment and GDP per
capita in 2008. See “Abbreviations” for a key to country abbreviations.
Figure 7.3 Distribution of Firms in ECA Region That Consider Worker Skills a
“Major” or “Very Severe” Constraint, 2008
9
mean = 30.2
8
TJK
7
UZB
number of countries
6
AZE ARM CZE
5
BIH GEO EST
4
KOS KGZ HRV UKR
3
MKD ALB LVA MDA
2
MNE SRB BGR POL LTU KAZ
1
HUN SVN TUR SVK ROM RUS BLR
0
%
%
%
10
20
40
50
60
70
80
90
30
x<
x<
x<
x<
x<
x<
x<
x<
x<
<
<
<
<
<
<
<
<
%
%
%
%
60
80
10
20
30
40
50
70
were constraining growth in ECA countries before the onset of the cur-
rent downturn. This is a puzzle that this book could not fully resolve,
mainly because crucial information is missing. International assessments
provide information about student competencies up to the age of 15 (usu-
ally the end of lower secondary education), but beyond that age, no infor-
mation exists on student competencies. Nor does information exist on the
quality of upper secondary or tertiary education or the relevance of educa-
tion at these levels. Moreover, no information exists on students’ noncog-
nitive (or socioemotional—“soft”—skills) (see World Bank 2011b for an
example of testing noncognitive skills in Peru). For policy makers to better
understand the role of education in contributing to skills bottlenecks—and
how to address it—this informational gap needs to be addressed.
Although the education offered in ECA countries at the end of lower
secondary is respectable relative to levels of income, it does not appear to
be adequate (or of the right relevance) to meet the demand for skills in the
region. In addition to higher levels of education, firms are demanding
higher-order skills that higher education alone does not guarantee, including
Extended Summary: The Path for Education Reforms in the ECA Region 199
cognitive skills such problem solving, together with behavioral skills such
as teamwork (see box 7.1). Such skills are increasingly a part of the land-
scape of work in modern economies. Figure 7.4, which illustrates findings
from the European Work Conditions Survey, demonstrates how the cogni-
tive demands of work have risen relative to more routine tasks.3
The need for higher-order skills is not, moreover, confined to a few
high-technology sectors, even construction—which is not normally viewed
as a highly skilled industry—requires the ability to learn on the job, work
independently, and solve complex problems. Nor are higher-order skills a
requirement only of upper-middle income countries. Surveys of employ-
ers in lower-middle income countries, such as Kazakhstan and the former
Yugoslav Republic of Macedonia, point in the same direction: the skills
that are lacking include communication, thinking skills, problem-solving
skills, and the ability to work independently (Ivaschenko 2008, World
Box 7.1
50
40
30
20
10
0
le en
th new
ta ng
ks
w ng
as
ob se
ex ki
of gi
s
gs
s
k
m
sk
or
st
pl rta
ds han
pr ore
ng
in
ou
m e
ni
f
co und
ho c
un
on
ar
et or
le
ot
g
m ng
in
on
lv
si
m
oo
so
g
ch
in
do
EU15 member states (i.e., non-ECA) new EU member states
Box 7.2
getting worse, not better. Rather than narrowing the gap between their
scores and OECD mean scores, many ECA countries that participate in
international learning assessments at this level—which are likely the
better-managed countries—either seem stuck or have regressed in terms
of their scores. It is probable, moreover, that the situation is even worse
in ECA countries that do not participate in efforts to measure and com-
pare the academic performance of their students. In fairness, the latest
round of PISA (2009) showed some improvements for 11 ECA countries,
202 Skills, Not Just Diplomas
Figure 7.5 Students Report Unofficial Payments Are “Usually” or “Always” Needed
in Public Technical Colleges and Universities, 2006
15
% of respondents
10
0
at er
ro rn
IS
CI
IC
b
Eu ste
es
pe
em
LI
M
Ea
m
st
h
EU
ut
So
usually always
these systems have been operating “in the dark” because they do not sys-
tematically collect data on the learning and employment outcomes of all
students. Second, the legacy of central planning has kept the governance
and management of school systems highly centralized, with central policy
makers deeply involved in operational details. While virtually all OECD
countries have embraced performance-oriented management in educa-
tion since the 1980s (see OECD 2005c)—albeit at different speeds—
most ECA countries continue to use management practices that focus on
compliance with detailed regulations and financing schemes based on
inputs, not outputs. This means that most local education authorities and
school principals in the ECA region lack the autonomy and authority to
make crucial management decisions for their own institutions, including
how much and what type of vocational content students can choose from
and how many teachers to employ. As a result, the education and training
system is inflexible and does not respond either to labor market needs or
to changes in student numbers.
Third, financial resources are ineffectually used in the region, particu-
larly in the pre-university sector, where few countries have adjusted teacher
staffing levels in response to falling student numbers over the past 20 years.
As a result, student-teacher ratios (STRs) have fallen sharply as per student
costs have risen—more so than in any other region in the world (see figure
7.6). This implies that scarce resources are misused by paying meager sala-
ries to too many staff members and heating half-empty buildings.
With few exceptions (notably Turkey), class sizes in ECA are also
smaller than those in other parts of the world. For instance, the average
class size at the primary level in Poland, Serbia, and Azerbaijan is 20, 14,
and 12, respectively, compared to 23 in France, 26 in the United Kingdom,
and 31 in Chile and the Republic of Korea (see figure 7.7). Although the
ECA region also has a large number of small schools, relatively small class
sizes are prevalent even in the largest schools, with the possible exception
of Armenia.
Small class sizes and lower STRs in the ECA region have not, how-
ever, improved educational quality—as the cost per student has
increased, student learning outcomes have not improved. Instead, these
conditions have led to poor teacher remuneration—especially at the
entry level, but also relative to other professions—and ultimately, to the
demoralization of the teaching workforce. This has discouraged talented
young students from becoming teachers, while simultaneously encourag-
ing those already in the profession to stay on as long as they can to
benefit from seniority-related pay. As a consequence, the profession as a
Extended Summary: The Path for Education Reforms in the ECA Region 205
29
27
pupil-teacher ratio (primary)
25
23
21
19
17
15
90
91
92
93
94
95
96
97
98
99
00
01
02
03
04
05
06
07
08
19
19
19
19
19
19
19
19
19
19
20
20
20
20
20
20
20
20
20
East Asia and Pacific Europe and Central Asia
Latin America and the Caribbean Middle East and North Africa
Sources: EdStats Database, plus authors’ linear interpolations for missing years.
Note: Actual data are marked with dots. One of the problems with international data on student-teacher ratios
(in both the World Bank’s EdStats Database and other international education databases) is that it is unclear
whether or not data for a particular country are reported on a full-time equivalent basis.
whole is failing to renew itself: few new recruits are entering the system
and the teaching cadre is aging (see figure 7.8). As the wage bill squeezes
out other spending, less resources are available for training and the training
received is often of questionable value. Given the importance of effec-
tive teaching, especially for inculcating higher-order cognitive skills,
ECA countries are thus poorly positioned to improve student learning
with their existing teaching workforce.4
To be fair, these impediments affect ECA countries differently and
vary accordingly to the level of education. However, no ECA country has
fully escaped the legacy of central planning.5 That system focused on
generating data on inputs—that is, it checked whether local actors were
in compliance with detailed norms for all inputs. Education ministries
remained in the dark, however, about the return on those inputs, that is,
whether students actually acquired skills and competencies.
In terms of how these impediments affect the different levels of educa-
tion, they are most clearly apparent at the pre-university level, which
absorbs two-thirds of total education funding in the ECA region. The
206
Figure 7.7 Comparison of Average Size of Primary School Classes Worldwide, Various Years
35
average class size (number of students)
30
25
20
15
10
0
Slovenia
Czech Republic
Estonia
Slovak Republic
Poland
Hungary
Bulgaria
Azerbaijan
Armenia
Russian Federation
Ukraine
Turkey
Serbia
Luxembourg
Greece
Spain
Finland
Germany
France
Ireland
United Kingdom
Mexico
United States
Brazil
Chile
Republic of Korea
OECD average
EU-11 LI CIS MI CIS Turkey Balkans Western Europe other
Sources: OECD Education At a Glance Database for OECD and partner countries; authors’ calculations based on Ministry of Education data (from education management information
systems) for Azerbaijan, Serbia, and Romania, and on Statistical Institute school-level data for Bulgaria.
Note: LI = low-income, MI = middle-income. Most recent data for each country (2005, 2006, 2007, or 2008).
Extended Summary: The Path for Education Reforms in the ECA Region 207
Figure 7.8 Percentage of Students with a Teacher Over 50 Years Old in ECA
Countries, Selected Years
60
% of children with teachers
50
over 50 years old
40
30
20
10
0
Bulgaria
Hungary
Lithuania
Romania
Slovak Republic
Slovenia
Macedonia, FYR
Turkey
Russian
Federation
Armenia
EU10+1 SEE MI CIS LI CIS
countries
1995 1999 2003 2007
Sources: IEA (1995, 1999, 2003, and 2007): Mathematics and Science Teacher Background Data Almanacs.
Note: LI = low-income, MI = middle-income.
Figure 7.9 Status of Measuring and Using Data on Student Learning Outcomes in
the ECA Region, 2009
Several years of
experience in
Several years of measuring,
experience with analyzing, and
own assessments making use of
and regular learning results to
Early piloting of participation in improve education
own assessment international policy
instruments, assessments
some participation
in international Bulgaria
Not started, very assessments
Hungary
early stages, no, or
rare participation Latvia
in international Lithuania
Albania
assessments Romania
Croatia
Armenia Serbia
Czech Republic
Azerbaijan Slovenia
Estonia
Belarus
Bosnia and Georgia
Herzegovina Macedonia, FYR
Kazakhstan Moldova
Kosovo Montenegro
Kyrgyz Republic Poland
Tajikistan Russian Federation
Turkey Slovak Republic
Turkmenistan Ukraine
Uzbekistan
Sources: Authors’ assessments based on data from UNICEF (2007, table 2.1); the extent of countries’ participation
in PISA, PIRLS, and TIMSS since 1995; and inputs from World Bank country experts. The UNICEF data is drawn
from “Table 2.1: Status of Reforms of Assessment and Examination Systems, 2006,” regarding “Introduction of
other school exams or assessments (e.g., basic school)” and “Introduction of sample-based national assessment.”
In this table, UNICEF scores the progress of countries on a range of 0 to 4, with 0 representing “not planned or
started” and 4 representing “operational.” In addition, the figure uses World Bank staff compilations on the
number of international assessments in which each country has participated.
example of the tracer study in the Netherlands (which has been in place
since 1989), table 7.1 provides an example of the types of information
such surveys can provide.
At the tertiary level, data on employment and learning outcomes are
also needed to shed light on how the sector is performing from a skills
perspective. Currently, data on tertiary outcomes (both in ECA and
around the world) focus on research outcomes, such as how many publi-
cations the faculty of a university is producing. Unfortunately, when this
is the only performance indicator available, it inadvertently receives more
attention than it should. Indeed, rankings of universities in the region
place significant weight on the number of publications that they produce.
However, from a skills perspective, the outcome that matters most is
whether students are graduating with the competencies in demand on
the job market and finding jobs. Currently, these outcomes are not being
measured, and, unfortunately, an international standardized learning
assessment for tertiary students is not expected to be launched by the
OECD until 2016 (see box 5.1). Thus, to gather the necessary data
now—when it is needed—ECA countries will need to begin with their
own, domestically developed approaches.
There are several complementary ways of providing more information
on the performance of the tertiary sector, some of which are more direct
than others. These include rankings or league tables, tracer studies, and
direct measurement of student competencies via standardized tests. One
way that ECA countries might begin to measure the learning outcomes
of tertiary students would be to introduce standardized testing within
fields of study where these outcomes have been clearly defined by the
National Qualification Frameworks which have been built in most coun-
tries (e.g., for graduates of nursing programs). Alternatively, ECA minis-
tries of education might define and measure a set of broad competencies
that all tertiary graduates are expected to have. The OECD’s DeSeCo
project (OECD 2005c), discussed in chapter 2, could provide a starting
point for the definition of such broad competencies.
Box 7.3
designed, such an approach could create exactly the wrong incentives for
improving learning. Rather, policy makers should use performance data
to identify both schools that need greater support and schools that offer
successful models for innovation.
Significant improvements in the direction of greater autonomy and
accountability will, however, be difficult to achieve in the school system
unless policy makers in the region address the demoralization of the
teaching force, an effort that will involve resolution of serious overstaffing
and improving the pay, professional development, and work conditions of
teachers.
Most countries in the region have shied away from adjusting teaching
staff levels in response to declining student numbers, even in contexts
where per student financing provides powerful incentives to reduce
teacher numbers. Addressing this inefficiency will free up funds for
more classroom learning materials. Judicious use of retirement incen-
tives for older teachers and stronger redundancy or retraining packages
214 Skills, Not Just Diplomas
Table 7.2 Tools for Strengthening Academic and Fiscal Integrity in University-Level
Institutions
Tools that strengthen academic integrity. . . Tools that strengthen fiscal integrity…
• Licensing • Financial audits
. . . and educational quality . . . and efficiency of spending
• Accreditation/academic audits/evaluations • Public disclosure laws
• Public disclosure laws • Strategic budget plans
• Fostering outsiders to review academic • Performance contracts based on
integrity (e.g., Romania) performance indicators
• Performance contracts based on perfor- • Student loans/scholarships/vouchers
mance indicators
• Creation and dissemination of rankings/ • Embedding incentives into allocated
benchmarks resources (e.g., tying funding to the
number of students enrolled or graduated,
rewarding good performers, etc.)
Source: Authors’ review of available instruments, as presented in Salmi (2009).
216 Skills, Not Just Diplomas
per student
financing widely
introduced
piloting of per
student financing
Armenia (2005)
Bulgaria (2008)
per student Kyrgyz Republic
financing under Czech Republic
(2006)
discussion (1992)
Moldova (2010) Estonia (2001)
Russian
Georgia (2007)
Azerbaijan Federation (1998)
not started or Hungary (1990)
Latvia Tajikistan (2005)
very early stages Kosovo (2002)
Macedonia, FYR Uzbekistan (2008)
in discussions Lithuania (2001)
Moldova
Poland (2000)
Serbia
Albania Romania (2010)
Slovenia
Belarus Slovak Republic
Bosnia and (2004)
Herzegovina
Croatia
Kazakhstan
Montenegro
Turkey
Ukraine
of funding for higher education and no single ideal mix of public and
private funding sources. Different countries will make different kinds
of trade-offs.
adult learning and create the tools needed to implement this strategy
(e.g., coordination mechanisms, plus initial steps toward regulation). Less
advanced economies in the region that are not facing a demographic
decline (i.e., low-income CIS countries and Albania) may find it more
productive to limit efforts to establishing a strategic policy framework
and coordination mechanism for this sector. For many countries, partici-
pating in the OECD’s Programme for International Assessment of Adult
Competencies (PIAAC) would be an important first step in understand-
ing the current skills and competencies of their work forces.
Summary
Reforming education systems in the ECA region so that they deliver
higher-quality and more relevant education to the majority of students
will go a long way towards addressing the skills gap that many countries
in the region face. Although the recent economic downturn has dimin-
ished the demand for labor and skills. Skills shortages will likely reemerge
once growth picks up again. The process of reforming education will take
time; consequently, ECA countries should not wait to begin.
Many needed changes—assessment systems that measure what stu-
dents learn, changes in the size and remuneration of the teaching force,
and strong quality assurance systems—are not without cost. Given that
economic growth is expected to be lower and fiscal constraints tighter in
the near to medium term, ECA countries will have little choice but to
free up resources that are currently funding existing inefficiencies in their
education systems, particularly those associated with school infrastruc-
ture and staff levels designed for much larger student populations. More
resources will also need to be raised from outside the education system,
especially at the tertiary level. Putting off the policy reform directions
discussed in this book—whether for financial or political reasons—risks
the deterioration of a fundamental source of national competitiveness: a
skilled workforce. This is clearly not an acceptable option for any country
in the region.
For movement of education systems away from an emphasis on inputs
and processes to an emphasis on improving the learning outcomes of the
majority of students, much more measurement of student learning out-
comes is needed at all levels of education. This data must then be used in
the policy process to identify strengths and weaknesses and help formu-
late appropriate policy responses. As the role of central governments is
Extended Summary: The Path for Education Reforms in the ECA Region 221
Notes
1. The PIRLS scale average is a metric established by setting the mean scores
of participating countries to 500 and the standard deviation to 100, thus
enabling comparisons over time, since all cycles are placed on this metric so
that scores are equivalent from cycle to cycle. In contrast, the international
average, obtained by averaging across the mean scores for each of the par-
ticipating countries, needs to be recomputed for each new cycle, based on
the set of participating countries and changes from cycle to cycle, depend-
ing on the set of countries taking part. For more information on PIRLS
2006, see its website at http://timss.bc.edu/pirls2006/index.html (accessed
January 2011).
2. The EBRD-World Bank Business Environment and Enterprise Performance
Surveys (BEEPS) provide successive snapshots of the self-reported constraints
to expansion by firms in the ECA region. The surveys were conducted in
virtually all ECA countries in 1999, 2002, 2005, and 2008. Survey samples
are constructed by random sampling from the national registry of firms (or
equivalent) and cover both industry and service sectors.
3. The Fourth European Work Conditions Survey conducted in 2005 covers the
EU-27 countries, Norway, Switzerland, Croatia, and Turkey (see Eurofound
2007).
4. The importance of effective teaching for learning is now increasingly recog-
nized. A good teacher can make a huge difference, with high-quality teachers
able to “get an entire year’s worth of additional learning out of their students
compared to those near the bottom” (Hanushek 2006, 3). That is, a good
Extended Summary: The Path for Education Reforms in the ECA Region 223
teacher will get a gain of 1.5 grade-level equivalents, while a bad teacher will
get only the equivalent of 0.5 years for a single academic year. When consid-
ered over the entire school cycle, such differences accumulate rapidly and
make the impact of other variables small in comparison.
5. Although Turkey comes from a different tradition, education management
remains highly centralized. The impediments identified in this section—
specifically, too little information on student skills and competencies and too
little autonomy and accountability—apply with equal measure to Turkey.
References
Abromovsky, Laura, Erich Battistin, Emla Fitzsimons, Alissa Goodman, and Helen
Simpson. 2005. The Impact of the Employer Training Pilots on the Take-up of
Training Among Employers and Employees. London: Department for Education
and Skills, Government of the United Kingdom.
Acemoğlu, Daron. 1998. “Why Do New Technologies Complement Skills?
Directed Technical Change and Wage Inequality.” Quarterly Journal of
Economics 113: 1055–89.
Acosta, Pablo, and Leonardo Gasparini. 2007. “Capital Accumulation, Trade
Liberalization, and Rising Wage Inequality: The Case of Argentina.” Economic
Development and Cultural Change 55 (July): 793–812.
Adelman, Clifford. 2003. “Learning Accountability from Bologna: A Higher
Education Primer.” Institute for Higher Education Policy, Washington, DC.
AES (Adult Education Survey) (dataset). 2007. Eurostat, Luxembourg. http://
epp.eurostat.ec.europa.eu/portal/page/portal/microdata/adult_education_
survey.
Airola, Jim, and Chinhui Juhn. 2005. “Wage Inequality in Post-Reform Mexico.”
IZA Discussion Paper 1525, Institute for the Study of Labor, Bonn.
Allais, Stephanie. 2010. The Implementation and Impact of National Qualifications
Frameworks: Report of a Study in 16 Countries. Geneva: International Labour
Organization.
Alam, Asad, Mamta Murthi, Ruslan Yemtsov, Edmundo Murrugarra, Nora
Dudwick, Ellen Hamilton, and Erwin Tiongson. 2005. Growth, Poverty, and
225
226 References
Inequality: Eastern Europe and the Former Soviet Union. Washington, DC:
World Bank.
———. 2010. “When They Come to School, I Will Teach Them: When Equality
Leads to Inequality.” English translation of a Serbian-language PowerPoint
presentation delivered to the Serbian Academy of Sciences and Arts confer-
ence, “The Change of Roman Identity, Culture, and Language Conditioned by
Planned Socioeconomic Integration,” Belgrade, December 6–8.
Alam, Asad, Paloma Anós Casero, Faruk Khan, and Charles Udomsaph. 2008.
Unleashing Prosperity: Productivity Growth in Eastern Europe and the Former
Soviet Union. Washington, DC: World Bank.
Alvarez, Jesus V., Garcia Moreno, and Harry A. Patrinos. 2007. “Institutional
Effects as Determinants of Learning Outcomes: Exploring State Variations in
Mexico.” Well-Being and Social Policy 3 (1): 47–68.
Amaral, Alberto, Glen A. Jones, and Berit Karseth, eds. 2003. Governing Higher
Education: National Perspectives on Institutional Governance. New York:
Springer.
Autor, David H., Lawrence F. Katz, and Melissa S. Kearney. 2005. “Trends in U.S.
Wage Inequality: Re-Assessing the Revisionists.” NBER Working Paper 11627,
National Bureau of Economic Research, Cambridge, MA.
Autor, David H., Frank Levy, and Richard J. Murnane. 2003. “The Skill Content of
Recent Technological Change: An Empirical Exploration.” Quarterly Journal
of Economics 118 (4): 1279–334.
Barnow, Burt S. 2009. “Vouchers in U.S. Vocational Training Programs: An
Overview of What We Have Learned.” Zeitschrift für Arbeitsmarktforschung
42 (1): 71–84.
Barrera-Osorio, Felipe, Tazeen Fasih, Harry Anthony Patrinos. 2009. “Decentralized
Decision-Making in Schools: The Theory and Evidence of School-Based
Management.” Report 48740, World Bank, Washington, DC.
Bartelsman, Eric, John Haltivanger, and Stefano Scarpetta. 2004. “Microeconomic
Evidence of Creative Destruction in Industrial and Developing Countries.”
Policy Research Working Paper 3464, World Bank, Washington, DC.
Bartlett, William. 2007. “Economic Restructuring, Job Creation, and the Changing
Demand for Skills in the Western Balkans Area.” In Labour Markets in the
Western Balkans: Challenges for the Future, ed. Anastasia Fetsi, 19–50. Turin:
European Training Foundation.
Baucal, Aleksandar. 2006. “Development of Mathematics and Language Literacy
of Roma Students.” Psihologija 39 (2): 211–35.
———. 2010. “When They Come to School, I Will Teach Them: When Equality
Leads to Inequality.” English translation of a Serbian-language PowerPoint
presentation delivered to the Serbian Academy of Sciences and Arts confer-
References 227
Contreras, D., L. Flores, and F. Lobato. 2003. “Monetary Incentives for Teachers
and School Performance: The Evidence for Chile.” Department of Economics,
University of Chile, Santiago.
Crostat (Republic of Croatia Central Bureau of Statistics). 2007. Raw dataset
from Croatia 2007 Labor Force Survey, Crostat, Zagreb, Croatia.
CVTS (Continuing Vocational Training Survey) (dataset). 1995, 2005. Eurostat,
Luxembourg. http://epp.eurostat.ec.europa.eu/portal/page/portal/education/
introduction.
Czech Republic, Government of. 2007. “The Strategy of Lifelong Learning in the
Czech Republic.” Government Decree 761/2007, Prague.
Dahlman, Carl. 2009. “Different Innovation Strategies, Different Results: Brazil,
Russia, India, China, and Korea (the BRICKs).” In Innovation and Growth:
Chasing a Moving Frontier, ed. Vandana Chandra, Deniz Eröcal, Pier C.
Padoan, and Carlos A. P. Braga, 131–68. Paris: Organisation for Economic
Co-operation and Development and World Bank.
Dar, Amit, Sudharshan Canagarajah, and Paud Murphy. 2003. “Training Levies:
Rationale and Evidence from Evaluations.” World Bank, Washington, DC.
Dearden, Lorraine, Leslie McGranahan, and Barbara Sianesi. 2004. An In-Depth
Analysis of the Returns to National Vocational Qualifications Obtained at Level
2. London: Centre for the Economics of Education.
Dearden, Lorraine, Howard Reed, and John van Reenen. 2000. “Who Gains
When Workers Train? Training and Corporate Productivity in a Panel of
British Industries.” CEPR Discussion Paper 2486, Centre for Economic Policy
Research, London.
Denhardt, Janet V., and Robert B. Denhardt. 2003. The New Public Service:
Serving, Not Steering. New York: M. E. Sharpe.
Duke, Chris, Abrar Hasan, Paul Cappon, Werner Meissner, Hilary Metcalf, and
Don Thornhill. 2008. OECD Reviews of Tertiary Education: Croatia. Paris:
Organisation for Economic Co-operation and Development.
EBRD (European Bank for Reconstruction and Development)–World Bank.
2006. “Life in Transition Survey 2006.” EBRD, London, and World Bank,
Washington, DC. Data available on the EBRD website, http://www.ebrd.com/
pages/research/analysis/publications/transition/data.shtml.
EC (European Commission). 2008. “New Skills for New Jobs. Anticipating and
Matching Labour Market and Skill Needs.” Commission Staff Working
Document accompanying the Communication from the Commission to the
European Parliament, the Council, the European Economic and Social
Committee, and the Committee of the Regions. EC, Brussels.
Edstats (database). World Bank, Washington, DC. http://go.worldbank.org/
ITABCOGIV1.
230 References
Ercan, Hakan. 2008. “The Evolution of Demand for Skills in Turkey.” Background
paper for Skills—Not Just Diplomas: Managing for Results in Education
Systems in Eastern Europe and Central Asia, ed. Lars Sondergaard and Mamta
Murthi (Washington, DC: World Bank, forthcoming). Unpublished; available
upon request.
Estermann, Thomas, and Terhi Nokkala. 2009. University Autonomy in Europe I.
Exploratory Study. Brussels: European University Association.
ETF (European Training Foundation). 2007. “Kyrgyzstan: ETF Country Plan
2007.” ETF, Turin.
EU (European Union). 2008. “Recommendation of the European Parliament
and of the Council of 23 April 2008 on the Establishment of the European
Qualifications Framework for Lifelong Learning.” Official Journal of the European
Union C (111): 1–7. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=
oj:c:2008:111:0001:0007:en:pdf.
Eurofound (European Foundation for the Improvement of Living and Working
Conditions). 2007. “Fourth European Working Conditions Survey” (EWCS
2005). Eurofound, Dublin, Ireland. The EWCS 2005 dataset is accessible
via the UK Data Archive (UKDA) of the University of Essex at www.esds
.ac.uk.
European Integration Consortium. 2009. “Labour Mobility within the EU in the
Context of Enlargement and the Functioning of the Transitional Arrangements.”
Employment, Social Affairs, and Equal Opportunities Directorate General,
European Commission, Nuremberg.
Eurostat SD (Statistics Database). Education and Training. Eurostat, Luxembourg.
http://epp.eurostat.ec.europa.eu/portal/page/portal/education/data/database.
———. Labour Market. Eurostat, Luxembourg. http://epp.eurostat.ec.europa.eu/
portal/page/portal/labour_market/introduction.
Fielden, J. 2008. “Global Trends in University Governance.” Education Working
Paper Series 9. Education Unit, Human Development Network, World Bank,
Washington, DC.
Flabbi, Luca, Stefano Paternostro, and Erwin R. Tiongson. 2007. “Returns to
Education in the Economic Transition: A Systematic Assessment Using
Comparable Data.” Policy Research Working Paper 4225, World Bank,
Washington, DC.
Fleisher, Belton M., Klara Sabirianova, and Xiaojun Wang. 2005. “Returns to Skills
and the Speed of Reforms: Evidence from Central and Eastern Europe, China,
and Russia.” Journal of Comparative Economics 33 (2): 351–70.
Glazerman, Steven, Eric Isenberg, Martha Bleeker, Amy Johnson, Julieta
Lugo-Gil, Mary Grider, and Sarah Dolfin. 2009. “Impacts of Comprehensive
Teacher Induction: Results from the First Year of a Randomized Controlled
References 231
Machin, Stephen, and John Van Reenen. 1998. “Technology and Changes in
Skill Structure: Evidence from Seven OECD Countries.” Quarterly Journal of
Economics 113 (4): 1215–44.
Mansoor, Ali, and Bryce Quillin, eds. 2007. Migration and Remittances: Eastern
Europe and the Former Soviet Union. Washington, DC: World Bank.
McKinsey & Company. 2007. “How the World’s Best-Performing School Systems
Come Out on Top.” McKinsey & Company, Social Sector Office, New York.
http://www.mckinsey.com/clientservice/Social_Sector/our_practices/
Education/Knowledge_Highlights.aspx.
———. 2008: “An Introduction to the McKinsey Report and What It Means
for Education in Australia.” Presentation, McKinsey & Company, New York.
Cited with permission.
———. 2010. “Closing the Talent Gap: Attracting and Retaining Top-Third
Graduates to Careers in Teaching—An International and Market Research-
Based Perspective” Byron Auguste, Paul Kihn, and Matt Miller, McKinsey
& Company, Social Sector Office, New York. http://www.mckinsey.com/
clientservice/Social_Sector/our_practices/Education/Knowledge_Highlights
.aspx.
Mitra, Pradeep, Marcelo Selowsky, and Juan Zalduendo. 2010. Turmoil at Twenty:
Recession, Recovery, and Reform in Central and Eastern Europe and the Former
Soviet Union. Washington, DC: World Bank.
Mungiu-Pippidi, Alina. 2009. “Assessing Corruption in the Romanian Higher
Education System: The Coalition for Clean Universities.” Photocopy, Hertie
School of Governance, Berlin.
Munich, Daniel, Jan Svejnar, and Katherine Terrell. 1999. “Returns to Human
Capital under the Communist Wage Grid During the Transition to a Market
Economy.” Review of Economics and Statistics 87 (1): 100–127.
Nastasescu, Vasile. 2006. “Romania: Higher Education System and Its Quality
Assurance.” Paper presented at The Third Europe and Central Asia (ECA)
Education Conference, “Quality and Relevance of Education,” St. Petersburg,
October 4–6.
Newell, Andrew, and Barry Reilly. 1997. “Rates of Returns to Educational
Qualifications in the Transitional Economies.” Discussion Paper in Economics
03/97, University of Sussex, Brighton.
Newell, Andrew, and Mieczyslaw W. Socha. 2007. “The Polish Wage Inequality
Explosion.” IZA Discussion Paper 2644, Institute for the Study of Labor,
Bonn.
OECD (Organisation for Economic Co-operation and Development). 1994. The
OECD Jobs Study: Facts, Analysis, Strategies. Paris: OECD.
———. 1999a. Measuring Student Knowledge and Skills: A New Framework for
Assessment. Paris: OECD.
References 235
———. 1999b. OECD Employment Outlook: Giving Youth a Better Start. Paris:
OECD.
———. 2004a. “Improving Skills for More and Better Jobs: Does Training Make a
Difference?” In OECD Employment Outlook 2004, 183–224. Paris: OECD.
———. 2004b. “Raising the Quality of Educational Performance at School.” Policy
Brief, OCED, Paris. http://www.oecd.org/publications/Pol_brief.
———. 2005a. Attracting, Developing, and Retaining Effective Teachers—Final
Report: Teachers Matter. Paris: OECD.
———. 2005b. “The Definition and Selection of Key Competencies—Executive
Summary.” OECD, Paris.
———. 2005c. Modernizing Government: The Way Forward. Paris: OECD.
———. 2005d. Promoting Adult Learning. Paris: OECD.
———. 2005e. “Thematic Review of Adult Learning: Poland Country Note.”
OECD, Paris.
———. 2006. “OECD Economic Surveys: Netherlands.” OECD, Paris.
———. 2007a. Analysis. Volume 1 of PISA 2006: Science Competencies for
Tomorrow’s World. Paris: OECD.
———. 2007b. Education at a Glance: OECD Indicators. Paris: OECD.
———. 2007c. Performance-Based Budgeting in OECD Countries. Paris: OECD.
———. 2007d. “PISA 2006: Science Competencies for Tomorrow’s World;
Executive Summary.” OECD, Paris.
———. 2008. Education at a Glance 2008: OECD Indicators. Paris: OECD.
OECD Education At a Glance Database. OECD, Paris. http://www.oecd.org/doc
ument/30/0,3343,en_2649_39263238_39251550_1_1_1_1,00.html.
OECD and World Bank. 2007. Reviews of National Policies for Education: Higher
Education in Kazakhstan. Paris: OECD.
Ok, Wooseok, and Peter Tergeist. 2003. “Improving Workers’ Skills: Analytical
Evidence and the Role of the Social Partners.” OECD Social, Employment,
and Migration Working Papers 10, Organisation for Economic Co-operation
and Development, Paris.
Orazem, Peter, and Milan Vodopivec. 1994. “Winners and Losers in Transition:
Returns to Education, Experience, and Gender in Slovenia.” Policy Research
Working Paper 1342, World Bank, Washington, DC.
Osborne, David, and Ted Gaebler. 1992. Reinventing Government: How the
Entrepreneurial Spirit is Transforming the Public Sector. Reading, MA: Addison-
Wesley Publishing Company, Inc.
OSI (Open Society Institute). 2007. Equal Access to Quality Education for Roma.
Vol. 1. Monitoring Report. Budapest: OSI/EU Monitoring and Advocacy
Program.
236 References
———. 2007. “From the Shortage of Jobs to the Shortage of Skilled Workers.
Labor Markets in the EU New Member States.” IZA Discussion Paper 3202,
Institute for the Study of Labor, Bonn.
———. 2008a. “Does the Labor Market Support Croatia’s EU Convergence?”
Photocopy, World Bank, Washington, DC. Unpublished background paper
for World Bank, 2009, “Croatia’s EU Convergence Report: Reaching and
Sustaining Higher Rates of Economic Growth.” Report 48879-HR, 2 vols.,
Europe and Central Asia Region, World Bank, Washington, DC.
———. 2008b. “Labor Market in Georgia: Lack of Jobs or Structural Mismatches?”
Photocopy, World Bank, Washington, DC. Unpublished background paper for
Georgia Country Economic Memorandum, World Bank, Washington, DC.
———. 2010. “Demand for Skills in FYR Macedonia.” Photocopy. Report sup-
ported by the Multidonor Trust Fund, “Labor Markets, Job Creation, and
Economic Growth,” through funding of the German, Norwegian, Austrian,
and Korean governments. World Bank, Washington, DC.
Rutkowski, Jan, and Stefano Scarpetta. 2005. Enhancing Job Opportunities: Eastern
Europe and the Former Soviet Union. With Arup Banerji, Philip O’Keefe,
Gaëlle Pierre, and Milan Vodopivec. Washington, DC: World Bank.
Salmi, Jamil. 2009. “The Growing Accountability Agenda in Tertiary Education:
Progress or Mixed Blessing?” World Bank Education Working Paper Series 16.
Document 47760, Education Unit, Human Development Network, World
Bank, Washington, DC.
———. 2011. “Review of Student Learning Assessment Experiences in Higher
Education.” Unpublished paper, World Bank, Washington, DC.
Santiago, Paulo, and Karine Tremblay. 2008. “Setting the Right Course, Steering
Higher Education.” With Marie-Claire Duguay and Thomas Weko. In Special
Features: Governance, Funding, Quality, Volume 1 of Tertiary Education for the
Knowledge Society, P. Santiago, K. Tremblay, E. Basri, and E. Arnal, 67–162.
Paris: OECD.
Scarpetta, Stefano, and Milan Vodopivec. 2005. “Restructuring, Productivity,
and Job Creation.” In Enhancing Job Opportunities: Eastern Europe and the
Former Soviet Union, ed. Jan J. Rutkowski and Stefano Scarpetta, 125–54.
Washington, DC: World Bank.
Silova, Iveta. 2009. “The Crisis of the Post-Soviet Teaching Profession in the
Caucasus and Central Asia.” Research in Comparative and International
Education 4 (4): 367–84.
Slantcheva, Snejana, and Daniel Levy, eds. 2007. The Search for Legitimacy:
Private Higher Education in Central and Eastern Europe. Basingstoke: Palgrave-
MacMillan.
Steiner-Khamsi, Gita, and Christine Harris-Van Keuren. 2008. “The Pendulum
of Decentralization and Recentralization Reforms: Its Impact on Teacher
238 References
Salaries in the Caucasus, Central Asia, and Mongolia.” With Iveta Silova
and Ketevan Chachkhiani. Background paper for Education for All Global
Monitoring Report 2009: Overcoming Inequality: Why Governance Matters.
Paris: UNESCO; London: Oxford University Press.
Steiner-Khamsi, Gita, and Iveta Silova. 2007. How NGOs React: Globalization
and Education Reform in the Caucasus, Central Asia and Mongolia. West
Hartfield, CT: Kumarian Press.
Stensaker, Bjørn, Jürgen Enders, and Harry de Boer. 2006. “Comparative
Analysis and Executive Summary.” Part One of “The Extent and Impact
of Higher Education Governance Reform Across Europe.” Final report
to the Directorate-General for Education and Culture of the European
Commission, Center for Higher Education and Policy Studies, University of
Twente, Enschede, The Netherlands.
Tan, Hong. 2001. “Do Training Levies Work? Malaysia’s HRDF and Its Effects on
Training, and Firm-Level Productivity.” Working Paper, World Bank Institute,
Washington, DC.
Tan, Hong, and Gladys Lopez Acevedo. 2005. “Evaluating Training Programs
for Small and Medium Enterprises, Lessons from Mexico.” Policy Research
Working Paper 3760, World Bank, Washington DC.
Tan, Hong, Yevgeniya Savchenko, Vladimir Gimpelson, Rostislav Kapelyushnikov,
and Anna Lukyanova. 2007. “Skills Shortages and Training in Russian Enter-
prises.” Policy Research Working Paper 4222, World Bank, Washington, DC.
Tumeneva, Yulia. 2006. “Analysis of Russian Education Features as Correlates
with Student Achievements in PISA.” In Russian Schools: From PISA 2000 to
PISA 2003, ed. Anatolii Kasprjak, 110–76. Moscow: Logos.
UIS (UNESCO Institute for Statistics). Data Centre (various databases).
UIS. Montreal. http://stats.uis.unesco.org/unesco/TableViewer/document
.aspx?ReportId=143&IF_Language=eng.
UNECE (United Nations Commission for Europe) SD (Statistical Database),
Labour Force and Wages. UNECE, Geneva, Switzerland. http://w3.unece
.org/pxweb/database/STAT/20-ME/3-MELF/?lang=1.
UNICEF (United Nations Children’s Fund). 2007. Education for Some More than
Others? Geneva: UNICEF.
United Kingdom, DFES (Department for Education and Skills). 2003. “The
Role of National Qualification Systems in Promoting Lifelong Learning.
Background Report for the United Kingdom.” Organisation for Economic
Co-operation and Development, Paris.
United Nations. 2006. “World Population Prospects—the 2006 Revision.”
Department of Economic and Social Affairs, Population Division, United
Nations, New York.
References 239
Usher, Alex. 2009. “Ten Years Back and Ten Years Forward: Developments and
Trends in Higher Education in Europe Region.” Paper presented at the
UNESCO Forum on “Higher Education in the Europe Region: Access, Values,
Quality and Competitiveness,” Bucharest, May 21–24.
Usher, Alex, and Jon Medow. 2009. “A Global Survey of University Rankings
and League Tables.” In University Rankings, Diversity, and the New Landscape
of Higher Education, ed. Barbara M. Kehm and Bjørn Stensaker, 3–18.
Vol. 18 of Global Perspectives on Higher Education. Rotterdam: Sense
Publishers. The university rankings of the Center for Higher Education
Development (CHE) in Gütersloh, Germany, can be accessed at http://www
.che-ranking.de/cms/?getObject=615&getLang=en.
Vegas, Emiliana, and Jenny Petrow. 2008. Raising Student Learning in Latin
America: the Challenge for the 21st Century. Washington, DC: World Bank.
Vincze, Maria, and Janos Kollo. 1999. “Economic Transformation and the Return
to Human Capital. The Case of Hungary, 1986–1996.” Budapest Working
Papers on the Labour Market 9907, Institute of Economics, Hungarian
Academy of Sciences, Budapest.
World Bank. 2003a. Raw dataset from 2003 Tajikistan Living Standards
Measurement Survey, World Bank, Washington, DC.
———.2003b. “Project Appraisal Document on a Proposed Loan in the Amount
of $60 Million to Romania for a Rural Education Project.” Report 25101-RO,
World Bank, Washington, DC.
———. 2003c. The World Development Report 2004: Making Services Work.
Washington, DC: World Bank.
———. 2005a. Expanding Opportunities and Building Competencies for Young
People. A New Agenda for Secondary Education. Washington, DC: World
Bank.
———. 2005b. Raw dataset from 2005 Uzbekistan Regional Panel Survey, World
Bank, Washington, DC.
———. 2007a. “Azerbaijan Efficiency Issues in General Education: Results from
a Public Expenditure Tracking and Service Delivery Survey (PETSDS) of
General Education in Azerbaijan.” Draft report, World Bank, Washington, DC.
———. 2007b. “Accelerating Bulgaria’s Convergence: The Challenge of Raising
Productivity.” Report 38570, Poverty Reduction and Economic Management
Unit, World Bank, Washington, DC.
———. 2007c. “The Labor Market and Education and Training in the Western
Balkans.” Draft policy note, World Bank, Washington, DC. http://siteresources
.worldbank.org/EXTECAREGTOPEDUCATION Resources/444607-11926
36551820/Labor_Market_and_Education_Note_Western_Balkans
.June_28a.pdf.
240 References
In fact, international test results show that many students—outside of a handful of coun-
tries in the region—are failing to acquire more than the most basic literacy and numeracy
skills. Anecdotal evidence also indicates that the rapid expansion in higher education has
led to a decline in the quality and relevance of education provided. At the same time, there
are few opportunities for adults to retrain, upgrade, or acquire new skills—the life-long
learning needed for employability.
As Skills, Not Just Diplomas suggests, the shortage of skills is a wake-up call to reform
education and training systems to provide higher quality education with the flexibility
for students and training institutions to better respond to market signals. Such deep
reform will have to center on the following:
• Focusing more on measuring whether students learn and graduates find jobs, and
using this information to actively improve teaching and learning.
• Using incentives across the education system, including granting greater autonomy
to institutions on curriculum, teaching methods, resource use and institutional mission,
and increasing accountability for learning.
• Addressing the huge inefficiencies in the sector by downsizing the school network
in response to dwindling student numbers, and using freed-up resources to restore
the attractiveness of the teaching profession, expand adult education and training, and
measure learning outcomes.
Only by embracing and seriously pursuing such difficult and long-term reforms will the
countries in Eastern Europe and Central Asia substantially improve their education outcomes
and support the competitiveness of their economic systems.
ISBN 978-0-8213-8096-3
SKU 18096